| name | gaia-intake-close |
| description | Post-merge intake closing skill. After a review/meta PR that closes intake issues is merged (or is CI-green and approved), run this skill to post standardized, personalized closing comments on the PR and each intake issue. Comments include: full evidence pipeline findings with /trust-appraise TM output, per-row artifact scores, decisions rationale, path-to-promotion guidance, and a badge status note. Use after /gaia-curate-chain completes and before or immediately after merge.
|
| version | 1.0.0 |
/gaia-intake-close
Posts standardized, human-readable closing comments on a merged (or approved) intake PR and its linked issues. Run this after the evidence pipeline and L4 review are complete.
What it posts
On the PR
A summary comment covering all skills in the PR:
- Full adversarial evidence audit findings table (per entry: type, verdict, action taken, reason)
/trust-appraise output with TM, grade, and per-row artifact scores- Final calibration decision for each named skill
- Badge status note (see below)
On each intake issue
A personalized closing comment addressed to the contributor (@handle) covering:
- Final decision (accepted/rejected/deferred)
- Calibrated star level and TM grade
- Which evidence entries were kept, corrected, or removed — and why
- Concrete path-to-promotion: what evidence type and threshold would push them to the next grade
- Link to the merged PR
Badge status note (always on PR, optionally on issue)
Explains the 2★ floor for badge generation, what unlocks at 2★, and links to gaiaskilltree.com/badges/ for self-service generation. If the contributor IS at 2★+, embed their actual badge SVG Markdown directly in the comment.
Inputs needed
Before running, have ready:
- The PR number (e.g.
1021)
- The list of intake issue numbers closed by the PR (e.g.
#677 #993)
- The
/trust-appraise --skill output for each new skill (run it inline if not already done)
- The adversarial audit findings (from
/ev-pipeline or inline adversarial review)
Procedure
Step 1 — Gather skill data
For each new skill in the PR, run:
PYTHONPATH=src python3 scripts/trust_appraise.py --skill contributor/skill-id
Read the named skill file to get final star level:
cat registry/named/<contributor>/<skill-id>.md | head -20
Step 2 — Fetch issue metadata
gh issue view <issue-number> --comments
Note the original contributor handle from the issue body (**User:** field in the batch summary).
Step 3 — Build the PR comment
Structure:
## Evidence Pipeline Findings — #<issue> [+ #<issue>...]
[For each skill:]
### #<issue> — `<skill-id>` (<contributor>)
**Final TM: X.X | Grade: <grade> | Calibrated: <N>★ (<status>)**
| Entry | Type | Verdict | Reason |
|---|---|---|---|
| ... | ... | KEPT/REMOVED/CORRECTED | ... |
### /trust-appraise
\`\`\`
TM: X.X Grade: <grade>
<type> score=X.X trust=XX <source>
\`\`\`
---
## Badges — Status for Newly Onboarded Contributors
[badge section — see below]
Step 4 — Build per-issue comments
Structure:
## Pipeline Review Complete — Final Decision
**Skill:** `<skill-id>`
**Contributor:** @<handle>
**Status: <Accepted/Rejected/Deferred> at <N>★ (<status>)**
### Evidence Pipeline Findings
[table]
### /trust-appraise
[output block]
### Path to <next level>
[concrete guidance: what evidence type, what threshold, example sources]
Thank you for the submission @<handle> — [one personalized sentence about the skill concept].
Step 5 — Badge section logic
if contributor_star_level >= 2:
post: "Your badges are live: "
else:
post: "Both contributors are at 1★ Awakened — below the 2★ floor for badge generation.
Visit gaiaskilltree.com/badges/ to generate your own once you reach 2★.
Path to 2★: [concrete guidance]"
Step 6 — Post comments
gh pr comment <pr-number> --body "..."
gh issue comment <issue-number> --body "..."
Evidence pipeline verdict taxonomy
Use these exact verdict labels in the findings table for consistency:
| Verdict | Meaning |
|---|
| KEPT | Entry passes all checks — URL live, metadata accurate, relevant |
| KEPT (corrected) | Entry kept but metadata was wrong — note what was fixed |
| REMOVED | Entry removed — note primary reason |
| DEFERRED | Entry not yet verifiable (e.g. SKILL.md not yet published) — note when it can be added |
Notes
- Always tag the contributor @handle in both the PR and issue comments — this builds transparency and trust
- Keep the "Path to promotion" section concrete: name the evidence type, the TM threshold, and an example source
- For
arxiv entries with 0 citations: always note "0 citations = 0 TM contribution per registry formula" so it's clear why they were removed/downgraded regardless of stated trust value
- For fabricated view counts: always note "view count not publicly available on source page" — do not guess or estimate
- The badge section is mandatory on every PR comment — it sets expectations for contributors who will look for their badge and find nothing
