菜单
Discovery-only global trending extension for Gaia. Snapshots configured external skill sources, prioritizes five-candidate pages for L4 review, and stops before trust, evidence, intake, or registry mutation.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Helper for release operations, compilation of changelogs between tags, version lockstep syncing, and promotion of GitHub releases from canary/pre-release to latest/production. Use when the user asks to "compile the changelog", "promote release to latest", "tag a release", or do version release operations.
Push committed changes and create a draft PR on GitHub. Use when you have staged changes ready to push and want to open a draft PR for early feedback or documentation.
Consult this skill BEFORE proposing a fix to any CLI gap or schema gap that touches evidence, trust magnitude, evidence types, evidence grades, the TM formula, or star-level promotion gates. Also read it whenever an issue mentions citations, stars, commits, contributors, views, reviewers, percentile, arxiv, github-stars-own, repo-own, social-signal, benchmark-result, peer-review, self-attestation, fusion-recipe, verifier-attestation, or proxy-containment — those are the 10 canonical evidence types and the raw inputs that feed their formulas. Also read it when asked whether a skill qualifies for a given star level (0★–6★): the Star Bar and promotion gates live in `META.md`, not in the TM methodology page. Reading both prevents proposing a calibration that satisfies the TM grade threshold but violates a Star Bar installability or link requirement.
Phase 4 of the Gaia evidence verification pipeline. Checks every URL in the evidence data lake for HTTP liveness — catches dead links (404, 403, timeout, connection errors) before evidence is ingested into the registry. Use this skill when: running the full /ev-pipeline, asked to "validate links", "check evidence URLs", "run link validation", "find dead links in the data lake", "check HTTP status of evidence sources", or "verify evidence links still work". Also useful standalone when you suspect evidence sources have gone stale after a long gap between pipeline runs. Wraps validate_sources.py, which uses Firecrawl CLI to scrape each unique URL and records pass/fail status in a markdown report. Requires Firecrawl CLI to be installed and authenticated.
Top-level orchestrator for the Gaia Registry evidence verification pipeline. Run this when you want to do a full evidence pass — collecting raw sources, verifying live GitHub star counts, adversarially auditing for quality/formatting issues, and checking that every URL is reachable — all in one coordinated sequence. Trigger phrases: "run the evidence pipeline", "full evidence pass", "verify evidence", "evidence verification pipeline", "run ev-pipeline", "check all evidence", "audit the data lake", "evidence quality sweep", "refresh evidence", "validate evidence sources". Also aliased as /evidence-verification-pipeline. Use the individual sub-skills (ev-collection, ev-star-verification, ev-adversarial-audit, ev-link-validation) only when you need to re-run one phase in isolation; for end-to-end work, always start here.
Canonical entry point for the full Gaia evidence verification pipeline. Invoke as /evidence-verification-pipeline or /ev-pipeline — both names trigger this skill. Use whenever you need to run all four evidence phases end-to-end: collecting raw evidence from the data lake, verifying live GitHub star counts, running adversarial auditing for noise and URL errors, and checking link health via Firecrawl. Also use when someone says "run the evidence pipeline", "verify the evidence lake", "audit evidence", "run ev-pipeline", "full evidence check", "pipeline run", "prepare evidence for ingestion", or "refresh the data lake". This is the pre-ingestion step — run it before importing any evidence into the registry.
| name | gaia-curate-trending |
| description | Discovery-only global trending extension for Gaia. Snapshots configured external skill sources, prioritizes five-candidate pages for L4 review, and stops before trust, evidence, intake, or registry mutation. |
| version | 1.0.0 |
| argument-hint | <source-manifest-or-run-id> |
Read ../gaia-curate/CURATION-CORE.md first. This ranks what reviewers should inspect now. It is external discovery, not Gaia's internal Trending API, not evidence, and not a trust signal. “Global” means every configured public source in the run manifest; never claim exhaustive web coverage.
Require a versioned manifest with bounded lanes and cursors:
contractVersion: curate-trending-manifest-v1
windowDays: [7, 30]
pageSize: 5
sources:
- id: source-id
lane: marketplace | source-repository | github-topic
url: https://...
cursor: null
Real marketplaces and source repositories are primary lanes. Package registries, MCP directories, model hubs, and extension listings are lead sources only: quarantine each lead until it resolves to an actually fetched upstream SKILL.md that passes the core artifact gate. Unknown sources become newSourceSuggestions; discovery must not edit registry/skill-sources.md.
Persist source-native observations for 7-day and 30-day windows: observed time, source rank/label, first-seen time, listing or download delta when exposed, repository star delta when exposed, and cross-source recurrence. Repository stars may prioritize inspection but are never inherited as implementation evidence or used for TM.
Assign one band mechanically in this priority order:
HOT: the source explicitly labels the artifact trending/top and it recurs on another configured source during the window.RISING: a comparable prior snapshot exists and at least one source-native rank/download/star delta is positive.NEW: first observed inside the selected window and neither higher band applies.STEADY: comparable snapshots exist with no positive trend signal.UNKNOWN: comparable history is missing, contradictory, or unavailable.Store the raw observations and rule that produced the band. Never ask a model to calculate the band and never convert it to evidence, grade, class, star level, TM, or acceptance.
Process at most five candidates per source page, and exactly one active candidate per model turn:
SKILL.md;DEFER; never infer acceptance.Allowed human/operator controls are NEXT, STOP, RESUME <run-id>, and one core decision for the active candidate. STOP writes a resumable checkpoint; it does not discard completed packets.
Write only under generated-output/curate-discovery/<run-id>/:
run.json
source-snapshots.json
candidates.jsonl
decisions.jsonl
deferred.jsonl
L4-REVIEW.md
L4-REVIEW.md lists candidate, trend band, source, existing-generic mapping, bounded disposition, and flags. Validate every row as discovery-packet-v1 and stop at L4. A shortlist decision is not intake, evidence, named-star, or mutation approval. A separate downstream workflow must perform those stages.