generate-cve-json

name	generate-cve-json
description	Generate a CVE 5.x JSON document from an <tracker> tracking issue, ready to paste into the Vulnogram `#source` tab of the ASF CVE tool at https://cveprocess.apache.org/cve5/<CVE-ID>#source. The conversion is deterministic: same issue in, same JSON bytes out. Handles multiple credits (one per line) and multiple references (URLs extracted from the issue's "Public advisory URL" and "PR with the fix" fields; the "Security mailing list thread" field is treated as internal-only and never exported).
when_to_use	Invoke when a security team member says "generate CVE JSON for NNN", "update the CVE tool entry for NNN", "paste-ready CVE for NNN", or is about to publish the advisory for a tracking issue and wants the CVE record filled in from the issue body in one paste-and-save step. Not appropriate before the CVE has been allocated (the script needs a CVE ID either from the issue body's `CVE tool link` field or from a `--cve-id` override).

generate-cve-json

This skill produces a CVE 5.x JSON document from a tracking issue in <tracker>, ready to paste into the Vulnogram "#source" tab of the ASF CVE tool. The goal is to eliminate the manual "copy each field from the issue into the right Vulnogram form input" step when you are preparing to publish an advisory.

Project-agnostic by design. All project-specific values (vendor, top-level product / package name, project display map, CNA org id, generator tag, …) are loaded from a TOML config the adopting project ships at <project-config>/tools/vulnogram/cve-json-config.toml. Concrete apache-foo-project-* strings appearing in this document are illustrative examples of how a project with a project-style package layout would configure things; replace them mentally with the adopter's own package taxonomy. The schema is documented in the package README.

Golden rule: the script generates a proposal JSON document. It parses a handful of structured fields from the issue body, but it cannot read the security team member's mind. Always review the generated JSON before pasting, and always do the final review inside Vulnogram before moving the CVE from DRAFT → REVIEW → READY → PUBLIC.

Determinism: the same input issue body produces exactly the same JSON bytes on every run. The script uses only the Python standard library, has no timestamps or machine-dependent values in its output, sorts JSON keys, and sorts references alphabetically. This lets you paste the result into Vulnogram, tweak fields in the tool, re-run the script later, and cleanly diff the two to see what the tool has added / what you changed by hand.

---

Inputs

• Issue number (required) — e.g. 232.
• Optional CLI overrides:
  • --cve-id CVE-YYYY-NNNN+ — override the CVE ID if the issue body's CVE tool link field has not yet been filled in, or retarget the JSON to a different CVE ID.
  • --title "<vendor>: <product>: …" — override the CVE title. Default is the GitHub issue title with the project's <vendor>: <product>: prefix (sourced from the TOML config) when it does not already start with that phrase.
  • --version-start X.Y.Z — override the start of the affected version range (the affected[].versions[].version field). Default is the lower bound parsed from the Affected versions field when it uses >= X, < Y syntax, otherwise "0".
  • --remediation-developer "Name" — append a type: "remediation developer" credit on top of whatever the body's Remediation developer field already lists (auto-populated by the security-issue-sync skill from the linked PR's author). Repeat the flag to add multiple developers; duplicates between the body field and CLI flags are dropped silently. The reporter credit(s) from the Reporter credited as field are always emitted with type: "finder".
  • --vendor / --product / --package-name / --collection-url — override the product identity fields. The defaults come from the project's TOML config (product.vendor, product.default_product, product.default_package_name, product.default_collection_url). They are used as the identity for Affected versions lines that don't start with a recognisable per-package directory name (see the multi-product note below).
  • --product-for PACKAGE=PRODUCT — override the CVE product display name for a specific packageName. Repeat to override multiple packages. Useful when a package is not in the project's project_display_map config, or when an acronym needs different casing from the title-cased fallback. Example: --product-for apache-foo-project-baz='Apache Foo Project Baz'.
  • --org-id <uuid> — override the CNA assigner org id (defaults to the ASF org id).
  • --discovery <word> — override source.discovery (default "UNKNOWN"; valid CVE 5.x values include UNKNOWN, INTERNAL, EXTERNAL, USER).
  • --no-envelope — emit only the inner cna container instead of the full CVE 5.x record (envelope is the default).
  • --attach — after generating the JSON, embed it at the end of the tracking issue's body (after the CVE tool link field), wrapped in a collapsible <details> block. The block is bracketed by HTML-comment markers (<!-- generate-cve-json: cve=CVE-YYYY-NNNN+ version=v1 --> … <!-- generate-cve-json:end cve=CVE-YYYY-NNNN+ version=v1 -->) that the script uses on later runs to find the existing block and replace it in place, so re-runs update the embedded attachment instead of duplicating it or breaking other body fields. The attachment lives in the body — not as a comment — so it stays above every status-change comment in the timeline (effectively "pinned" without needing any pin mechanism). Requires the positional issue argument; incompatible with --stdin.

---

Prerequisites on the tracking issue

For the generated JSON to be useful, the issue body should already be filled in through a prior security-issue-sync run. In particular:

• Short public summary for publish — becomes the CVE description.

• Affected versions — becomes the CVE affected[] list. The script understands the common version-expression shapes (< 3.2.2, >= 2.0.0, < 3.2.2, <= 3.2.1, a bare version like 3.1.5, and a bare lower bound like >= 2.0.0). Multi-product CVEs are supported — put one package per line, prefixing each with the package directory name as it appears in the adopter's repo, and the script emits one affected[] entry per line with the right product / packageName. Example (illustrative — using a hypothetical apache-foo project's sub-project layout):

  apache-foo-project-alpha <=6.5.0
  apache-foo-project-beta <=1.9.0
  

  Known package directory names are resolved to the vendor-preferred display casing via the project's packages.project_display_map config table; unknown packages fall back to title-cased dash-split and can be overridden with --product-for. A line without a package prefix (or a single-line field) falls back to the --product / --package-name defaults, which preserves the single-product behaviour.

  < NEXT VERSION placeholder — multi-package trackers don't know which package version will ship the fix until the wave's release manager picks it during a release cut. Until then, the Affected versions lines use the literal token NEXT VERSION as the upper bound, e.g.:

  apache-foo-project-alpha < NEXT VERSION
  apache-foo-project-beta < NEXT VERSION
  

  The generator strips < NEXT VERSION before parsing each line and emits a versions[] entry without lessThan (open-ended upper bound — "affected from <low> onwards, no fix released yet"). When the wave ships and the version is known, the security-issue-sync skill replaces each NEXT VERSION with the actual < X.Y.Z and the next regen produces a fully-bounded entry. Case-insensitive; combines with a lower bound (e.g. >= 2.0.0, < NEXT VERSION becomes {version: "2.0.0", status: "affected"}).

• Security mailing list thread — internal navigation reference only; the script does not export URLs from this field into references[]. Keep whatever the reporter or triager put there.

• Public advisory URL — each URL in this field is extracted and added to references[] with tags: ["vendor-advisory"]. Populated by the release manager (or the security-issue-sync skill) once the advisory is archived on <users-list>. The --advisory-url CLI flag still exists for ad-hoc overrides.

• PR with the fix — each URL in this field becomes a reference URL. Multiple URLs are supported: paste them on separate lines, as a bullet list, or comma-separated — the script extracts every https?://… token it finds.

• Reporter credited as — each line becomes one CVE credit entry with type: "finder". Multiple credits are supported: put each person on their own line. Full Name, Affiliation on a single line is treated as one credit, not two, so the common Jed Cunningham, Astronomer pattern works as expected. Bullets (- , * , 1. ) are stripped. Blank lines are ignored. If you need to credit many people::

  Jed Cunningham
  Saurabh Banawar
  selen (Huntr bounty 3e88d364-5047-4768-a52c-6568f21ef35b)
  

• Remediation developer — each line becomes one CVE credit entry with type: "remediation developer". Same parsing rules as Reporter credited as (newline-separated, Full Name, Affiliation is one credit, bullets stripped). Auto-populated by the security-issue-sync skill from the linked PR's author the first time PR with the fix is set; manual edits survive subsequent syncs (the skill only proposes appending names that aren't already there). The --remediation-developer CLI flag adds further names on top of whatever the body already lists.

• CWE — CWE-285: Improper Authorization style works; so does a bare CWE-285 or a plain sentence. The script extracts the CWE-\d+ token for the cweId field and uses the rest as the human-readable description.

• Severity — None, Low, Medium, High, Critical (case-insensitive) are emitted as the text content of a metrics[].other block. Vulnogram lets you replace this with a CVSS vector in its form if you want a numeric score.

• CVE tool link — the ASF CVE tool URL, e.g. https://cveprocess.apache.org/cve5/CVE-2026-40913. The script extracts the CVE-YYYY-NNNN+ token from this field. If the field is still _No response_, pass the CVE ID with --cve-id.

If one of these fields is missing, the JSON still generates, but the reviewer will need to fill the gap in Vulnogram. The skill surfaces any empty field in the proposal so nothing is silently skipped.

---

Prerequisites

• gh CLI authenticated with collaborator access to <tracker> — the script reads the tracker via gh.
• uv installed — the script is a small uv-managed Python project and is invoked as uv run --project tools/vulnogram/generate-cve-json generate-cve-json <N>.

See Prerequisites for running the agent skills in README.md.

---

Step 0 — Pre-flight check

Before reading the tracker:

1. gh api repos/<tracker> --jq .name returns the adopter's tracker repo name (per <project-config>/project.md), and
2. uv --version returns.

If either fails, stop and tell the user what to install or log in to.

---

Step 1 — Verify the issue has the required fields

Fetch the issue body and check every template field the script reads. If a field is missing or still _No response_, either run security-issue-sync first to fill it in, or override it on the command line.

gh issue view <N> --repo <tracker> --json body --jq .body \
  | grep -E '^###|^_No response_'

Ask the user whether to proceed if any critical field is empty (description, affected versions, CVE tool link, credits). Do not silently generate a JSON with placeholder values.

---

Step 2 — Run the generator

Run the project's console script through uv run --project, which prepares the (cached) virtualenv on first use and reuses it on later runs:

uv run --project <framework>/tools/vulnogram/generate-cve-json generate-cve-json <N> \
  --output /tmp/<CVE-ID>.json \
  --version-start <earliest-affected-version>

--version-start is the one flag the tracking issue body almost never contains and that Vulnogram expects filled in (the body field usually encodes only the upper bound). The remediation developer credit comes from the body's Remediation developer field, populated by the security-issue-sync skill from the linked PR's author — no CLI flag needed in the normal flow. For a fix that landed in 3.2.2 and was first introduced in 3.0.0, for example:

uv run --project <framework>/tools/vulnogram/generate-cve-json generate-cve-json 232 \
  --output /tmp/CVE-2026-40913.json \
  --version-start 3.0.0

Pass --remediation-developer "Name" only when you need to add a developer credit on top of (or in place of) what the body already contains — for example a co-author who didn't end up as the PR's GitHub author.

Additional flags, all optional:

• --cve-id CVE-YYYY-NNNN+ — override the CVE ID if the CVE tool link field is empty.
• --title "<vendor>: <product>: …" — override the title.
• --vendor / --product / --package-name / --collection-url — override product identity (defaults sourced from the project's TOML config under [product]).
• --org-id <uuid> — override the CNA assigner org id (defaults to the ASF org id).
• --discovery UNKNOWN|INTERNAL|EXTERNAL|USER — override source.discovery.
• --no-envelope — emit only the cna container (no cveMetadata, no dataType/dataVersion wrapper). Use this if Vulnogram's #source tab is in "inner block only" mode.
• --stdin — read the issue body from stdin instead of calling gh. Useful for offline iteration and for drafting by hand.

The script is deterministic — re-running it with the same flags and the same tracking-issue body produces the same JSON bytes.

Output shape (in brief)

The generated record matches what Vulnogram exports after a save, minus editor cruft. Notable fields:

• containers.cna.affected[] — vendor, product, collectionURL, packageName and a versions[] entry with version, lessThan, status: "affected", versionType: "semver".
• containers.cna.descriptions[] — both a plain value and an HTML supportingMedia alternative (Vulnogram's WYSIWYG mode needs both).
• containers.cna.problemTypes[].descriptions[] — cweId, human-readable description, type: "CWE".
• containers.cna.metrics[].other — type: "Textual description of severity" and content.text = the severity word.
• containers.cna.credits[] — one entry per Reporter credited as line (type "finder"), plus one entry per Remediation developer body line and per --remediation-developer CLI override (type "remediation developer"); duplicates between the body field and CLI flags are dropped silently.
• containers.cna.references[] — URLs with auto-tagged tags:
  • GitHub pull/ or commit/ URLs → ["patch"];
  • lists.apache.org / security.apache.org → ["vendor-advisory"];
  • everything else → no tag.
• containers.cna.source.discovery — "UNKNOWN" by default.
• containers.cna.providerMetadata.orgId — ASF assigner org id.
• cveMetadata — assignerOrgId, cveId, serial, state: "PUBLISHED".

---

Step 3 — Surface the output to the user

After the script finishes, print these three things in order:

1. The output file path, with a one-line cat suggestion so the user can review the JSON in the terminal:

   ```
   Wrote /tmp/cve-CVE-2026-40913.json
   cat /tmp/cve-CVE-2026-40913.json
   ```
   

2. A clipboard-copy command appropriate to the user's platform. On Linux with xclip installed:

   ```
   xclip -selection clipboard < /tmp/cve-CVE-2026-40913.json
   ```
   

   On Wayland: wl-copy < /tmp/cve-...json. On macOS: pbcopy < …. If xclip / wl-copy / pbcopy is not on PATH, skip the clipboard command and tell the user to copy manually.

3. The Vulnogram #source paste URL, as a clickable link rendered per the "Linking CVEs" rule in AGENTS.md:

   ```
   Paste the JSON into the Vulnogram #source tab:
     [CVE-2026-40913](https://cveprocess.apache.org/cve5/CVE-2026-40913#source)
   ```
   

The #source tab on the ASF CVE tool is the direct "paste raw JSON" view of the Vulnogram form. The page loads the current record, you paste the script output over the top, click Save, and the form view reflects the new values.

Optional: --attach to embed (or refresh) the JSON in the issue body

If the user also wants the JSON attached to the tracking issue itself (so it is discoverable from the issue without needing the local file), add --attach to the invocation:

uv run --project <framework>/tools/vulnogram/generate-cve-json generate-cve-json 232 \
  --output /tmp/CVE-2026-40913.json \
  --version-start 3.0.0 \
  --attach

What --attach does:

• After generating the JSON (exactly the same bytes as without --attach), edits the tracking issue's body to embed the full JSON inside a four-backtick fenced code block, collapsed behind a <details> disclosure so long records don't bloat the issue view. The block is appended after the existing template fields, right after the CVE tool link field, so it lives at the end of the body.
• Brackets the attachment with a pair of hidden HTML-comment markers (<!-- generate-cve-json: cve=CVE-YYYY-NNNN+ version=v1 --> … <!-- generate-cve-json:end cve=CVE-YYYY-NNNN+ version=v1 -->) so subsequent runs can find the existing embedded block and replace it in place, without spawning duplicates and without touching any other text in the body.
• Re-running with --attach is safe and idempotent: same issue body → same JSON → the script patches the body, leaving you with one and only one embedded attachment per CVE id. If the current body already matches what the script would write, the PATCH is skipped entirely (no no-op timestamp on the issue).
• The script prints Embedded CVE JSON in issue body on <tracker>#NNN on first run and Replaced CVE JSON in issue body on <tracker>#NNN on subsequent runs, followed by a URL that deep-links to the ## CVE JSON — paste-ready for … heading anchor inside the body.

Why embedded in the body and not as a comment? Two reasons:

1. Natural "pinning" without an API for it. GitHub has no pin-comment API. A separate comment ends up buried below every status-change comment — so a newcomer looking at the issue sees a long comment timeline with no obvious way to find the CVE JSON. The issue body always renders above the entire comment timeline, so anything embedded in the body is effectively pinned.
2. One place to read the tracker. The reporter-template fields, the CVE metadata, and the paste-ready JSON are all in one place — no hunting through the timeline to reconstruct the current state.

(GitHub also does not expose its user-attachments file-upload pipeline to the REST API — only the web UI drag-and-drop uses it — so a real file attachment isn't available to automation anyway. Embedding as body text is the closest automatable equivalent and is directly visible without a download round-trip.)

Confidentiality. The embedded block lives inside the private repo, so it inherits the repo-wide confidentiality rules. Linking CVE references inside the block follows the "Linking CVEs" rule in AGENTS.md: before publication the block links the ASF CVE tool; after publication, re-running the script includes a cve.org link as well.

---

Step 4 — Propose an issue comment recording the update

Per the "Keeping the reporter informed" rule in README.md, any status change on an issue must be recorded in an issue comment. Pasting a new version of the CVE record is a status change. Propose (and, on confirmation, post) a short comment like:

CVE entry regenerated from the tracking issue — generated paste-ready JSON for CVE-2026-40913 from the current body fields (description, affected versions < 3.2.2, CWE-285, Low severity, N credits, M references). Pasted into the Vulnogram #source tab; the record is now in sync with the tracking issue.

Include a count of credits and references so a later reviewer can sanity-check that nothing was dropped.

---

Step 5 — Never edit the JSON in place after pasting

Once the JSON has been pasted into Vulnogram and saved, do not edit the local JSON file to match tool-side changes. Re-run the script instead (it is deterministic — you will get a clean baseline), diff the new output against the current Vulnogram state, and paste the merged JSON back. This keeps the tracking issue as the single source of truth: if Vulnogram shows something different from the generated JSON, either the issue body is out of date and needs a security-issue-sync run, or the tool-side difference is intentional and the reviewer will keep it.

---

Guardrails

• Confidentiality. The script deliberately drops any URL that points at cveprocess.apache.org or the project's <tracker> repo from the references list before serialising. Those URLs are private ASF-internal links and should not appear in a published CVE record. See the "Confidentiality of <tracker>" section of AGENTS.md.
• CVE IDs are always linked per the "Linking CVEs" rule in AGENTS.md. When the skill mentions the CVE in proposals, recaps or comments on the <tracker> issue, it must render the ID as a markdown link — before publication to the ASF CVE tool, and additionally to cve.org after publication.
• <tracker> references are always linked per the "Linking <tracker> issues and PRs" rule in AGENTS.md. When the skill mentions the tracking issue in its own comments, render it as a markdown link.
• Deterministic output is a feature. Do not introduce timestamps, random UUIDs, ordering dependencies on dict iteration, or other sources of non-determinism into the script. If you need to add a new field, make sure the output still hashes the same across runs on the same input.
• Multi-entry fields. Credits are split on newlines only to preserve the Full Name, Affiliation pattern. References are extracted from URL tokens in the field value. Do not reintroduce comma-splitting on credits.
• No envelope means no metadata. --no-envelope drops the cveMetadata block which includes the CVE ID; the JSON is pure CNA content. Make sure the user knows they will have to set the CVE ID by hand in Vulnogram in that mode.

---

References

• AGENTS.md — repo-wide conventions (confidentiality, Linking CVEs, Linking <tracker> issues and PRs, release-branch defaults).
• README.md — handling process, in particular step 13 (fill in CVE tool fields and send advisory from the tool) and step 15 (paste the attached JSON into Vulnogram's #source tab, move the CVE to PUBLIC, close the issue).
• security-issue-sync — the sibling skill that populates the tracking issue fields this skill consumes.
• security-issue-fix — the other sibling skill that opens a public PR and updates the tracking issue with the fix URL.