Skip to content

gaia-research/gaia-skill-tree / .agents/skills/gaia-intake-close

gaia-intake-close

Verified

Post-review intake closing skill. After an intake is resolved (whether accepted via review/meta PR merge or rejected/deferred at triage/evidence review), run this skill to post standardized, personalized closing comments on the PR (if any) and each intake issue. Comments include: evidence pipeline findings with /trust-appraise TM output, per-row artifact scores, decisions rationale, path-to-promotion/acceptance guidance, and a mandatory badge status note including the universal Powered by Gaia badge. Use after curation completes, before or immediately after merge for accepted intakes, or immediately upon rejection.

/gaia-intake-close

Posts standardized, human-readable closing comments on an intake issue and its linked PR (when accepted) or directly on the intake issue (when rejected or deferred). Run this after the evidence pipeline and review gates are complete, or immediately upon a triage/review rejection decision.

What it posts

On the PR (Accepted intakes only)

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 with live rendered badges for newly onboarded contributors at ≥2★
  • Universal Powered by Gaia badge snippet

On each intake issue (All outcomes: Accepted, Rejected, Deferred)

A personalized closing comment addressed to the contributor (@handle) covering:

  • Final decision (accepted/rejected/deferred)
  • Calibrated star level and TM grade (for accepted), or rejection/deferral basis (for rejected/deferred)
  • Which evidence entries were kept, corrected, or removed — and why (or triage findings if rejected before evidence review)
  • Concrete path-to-promotion (for accepted) or path-to-acceptance (for rejected/deferred): what changes or evidence types would clear the bar
  • Mandatory Badge Status Note: Explains the badge status for the contributor, embeds live badges (for accepted ≥2★), explains the 2★ floor and links to gaiaskilltree.com/badges/, AND always includes the universal Powered by Gaia badge snippet
  • Link to the merged PR (if accepted)

Badge status note (Mandatory on every closing comment)

Every closing comment — whether accepted, rejected, or deferred — must include a ### Badges section:

  • Universal Powered by Gaia badge (Every submitter gets this): Anyone submitting an intake to Gaia is entitled to display the universal Powered by Gaia badge in their repository, even if their submission was rejected or deferred:
    [![Powered by Gaia](https://gaiaskilltree.com/badges/powered-by-gaia.svg)](https://gaiaskilltree.com)
    
  • Accepted at ≥2★: In addition to the universal badge, embed their actual named skill badge SVG image using Markdown image syntax (![alt](url) — never a bare link), provide a copy-paste Markdown embed snippet for their README, and link to self-service generation at gaiaskilltree.com/badges/?u=<handle>&s=<skill>.
  • Accepted at 1★ (below 2★ floor): Explain the 2★ Named floor for skill/rank badge generation ("1★ skills exist, 1★ badges do not"), provide the universal Powered by Gaia badge, and link to gaiaskilltree.com/badges/.
  • Rejected or Deferred: Explain that named skill and rank badges are reserved for accepted skills that reach the 2★ Named floor with verified evidence, provide the universal Powered by Gaia badge for their repo, and explain how resubmission reaching 2★ unlocks named badges.

Inputs needed

Before running, have ready:

  • The intake issue number(s) (e.g. #1793, #1766)
  • The outcome: accepted, rejected, or deferred
  • If accepted: The PR number (e.g. #1800), /trust-appraise --skill output, calibrated star level, adversarial audit findings
  • If rejected/deferred: The triage/audit findings, specific reasons (e.g. non-generalizable, missing attribution, unresolvable evidence), and constructive path to acceptance

Procedure

Step 1 — Gather skill data (Accepted intakes)

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

(For rejected intakes, gather the rejection reasons and actionable generalization/attribution guidance instead.)

Step 2 — Fetch issue metadata

gh issue view <issue-number> --comments

Note the original contributor handle from the issue body (**User:** or contributor: field in the batch summary). Check if the contributor already has existing badges:

ls docs/badges/_assets/<handle>/ 2>/dev/null

Step 3 — Build the PR comment (Accepted intakes)

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: score=X.X trust=XX


---

## Badges — Status for Newly Onboarded Contributors

@<handle> is at <N>★ — badges are live now:

![@<handle>/<skill-id>](https://gaiaskilltree.com/badges/_assets/<handle>/<skill-id>.svg)

Visit `gaiaskilltree.com/badges/?u=<handle>&s=<skill-id>` to generate the seal variant and any additional badge types.

Universal **Powered by Gaia** badge:
```markdown
[![Powered by Gaia](https://gaiaskilltree.com/badges/powered-by-gaia.svg)](https://gaiaskilltree.com)

### Step 4 — Build per-issue comments

#### Variant A: Accepted Intake
```markdown
## Pipeline Review Complete — Final Decision

**Skill:** `<skill-id>`
**Contributor:** @<handle>
**Status: Accepted at <N>★ (<status>)**

### Evidence Pipeline Findings
[table]

### /trust-appraise
[output block]

### Path to <next level>
[concrete guidance: what evidence type, what threshold, example sources]

### Badges
[badge section — see Step 5]

---

Thank you for the submission @<handle> — [one personalized sentence about the skill concept].

Merged in #<pr>.

Variant B: Rejected / Deferred Intake

## Pipeline Review Complete — Final Decision

**Skills:** `<skill-id-1>`, `<skill-id-2>`
**Contributor:** @<handle>
**Status: Rejected** (or Deferred) — [brief summary, e.g. "no evidence pipeline was run since this didn't clear intake triage; nothing to calibrate"]

### Findings

| Skill | Verdict | Reason |
|---|---|---|
| `<skill-id>` | REJECTED / DEFERRED | [Specific, falsifiable reason: e.g. tool dispatch without attribution, product-specific coupling] |

### Path to acceptance — `<promising-skill-id>`

[Actionable, constructive guidance: e.g. generalize the capability name and flow, credit upstream author, provide reproducible benchmarks]

### Badges
[badge section for rejected/deferred — see Step 5]

---

Thanks again for the batch, @<handle> — [encouraging closing sentence acknowledging effort and inviting revised resubmission].

Step 5 — Badge section logic (Mandatory for ALL comments)

if outcome == "accepted" and contributor_star_level >= 2:
    # Contributor is at or above the 2★ badge floor — EMBED LIVE BADGE IMAGE + POWERED BY GAIA
    # CRITICAL: Always use Markdown image syntax ![alt](url) — NEVER a bare link [text](url)
    post: """### Badges

`<contributor>/<skill-id>` is at <N>★ (<status>), so your badges are live:

![@<contributor>/<skill-id>](https://gaiaskilltree.com/badges/_assets/<contributor>/<skill-id>.svg)

Embed in your `README.md`:
```markdown
[![Gaia](https://gaiaskilltree.com/badges/_assets/<contributor>/<skill-id>.svg)](https://gaiaskilltree.com/u/<contributor>/)

Self-service badge generation, rank badge, and seal variants are at gaiaskilltree.com/badges/?u=&s=.

You can also use the universal Powered by Gaia badge:

Image: Powered by Gaia

[![Powered by Gaia](https://gaiaskilltree.com/badges/powered-by-gaia.svg)](https://gaiaskilltree.com)
```"""

elif outcome == "accepted" and contributor_star_level < 2:
    # 1★ Awakened — below the 2★ badge floor, but gets Powered by Gaia badge
    post: """### Badges

`<contributor>/<skill-id>` is at 1★ Awakened — below the 2★ Named floor for skill-specific badge generation ("1★ skills exist, 1★ badges do not").

However, you can use the universal **Powered by Gaia** badge in your repository right now:

[![Powered by Gaia](https://gaiaskilltree.com/badges/powered-by-gaia.svg)](https://gaiaskilltree.com)

```markdown
[![Powered by Gaia](https://gaiaskilltree.com/badges/powered-by-gaia.svg)](https://gaiaskilltree.com)

Skill-specific badges (handle, rank, <skill-id>) unlock at 2★ Named. Visit gaiaskilltree.com/badges/ to preview them. Path to 2★: [concrete guidance on evidence needed to reach TM 20]."""

else: # Rejected or Deferred intake — gets the universal Powered by Gaia badge post: """### Badges

Skill-specific and rank badges unlock at 2★ Named with verified evidence, so no named skill badge was generated for this batch. However, as an intake participant, you can use the universal Powered by Gaia badge in your repository right now:

Image: Powered by Gaia

[![Powered by Gaia](https://gaiaskilltree.com/badges/powered-by-gaia.svg)](https://gaiaskilltree.com)

Once a generalized resubmission addressing the findings above is accepted and calibrated to 2★+, automated badges (skill badge, rank badge, and seal variants) will unlock and become live at gaiaskilltree.com/badges/."""


### Step 6 — Post comments and close issues

**For Accepted Intakes:**
```bash
gh pr comment <pr-number> --body-file /tmp/pr-close-comment.md
gh issue comment <issue-number> --body-file /tmp/issue-close-comment.md
gh issue close <issue-number>

For Rejected Intakes:

# Ensure intake:rejected is applied
gh issue edit <issue-number> --add-label "intake:rejected"
# Post the comprehensive closing comment
gh issue comment <issue-number> --body-file /tmp/issue-close-comment.md
# Close the intake issue
gh issue close <issue-number>
# Clean up any staged review branch if pushed
git push origin --delete review/meta/<handle>--<skill> 2>/dev/null || true

Evidence pipeline verdict taxonomy

Use these exact verdict labels in the findings table for consistency:

VerdictMeaning
KEPTEntry passes all checks — URL live, metadata accurate, relevant
KEPT (corrected)Entry kept but metadata was wrong — note what was fixed
REMOVEDEntry removed — note primary reason
DEFERREDEntry not yet verifiable (e.g. SKILL.md not yet published) — note when it can be added
REJECTEDEntry or skill rejected — does not meet criteria (unattributed, product-coupled, etc.)

Notes

  • Always tag the contributor @handle in both the PR and issue comments — this builds transparency and trust
  • Keep the "Path to promotion" / "Path to acceptance" section concrete: name the evidence type, the TM threshold, or specific generalization steps
  • The ### Badges section is MANDATORY on EVERY closing comment — accepted, rejected, and deferred alike. Contributors look for their badges or want to know what unlocks them; never leave them guessing
  • Every intake participant receives the universal Powered by Gaia badge (powered-by-gaia.svg), including rejected and 1★ submissions. Never omit it
  • Always format live badges as Markdown images ![alt](url), NEVER bare hyperlinks [text](url). Bare links do not render the SVG visual in GitHub comments
  • 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
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 322kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall