Deliverable Selfloop

What the source does

• FilesystemNone observedNo reads or writes in the source
• NetworkNone observedNo outbound calls in the source
• ShellNone observedNo commands run in the source
• DependenciesPinnedNothing floats
• SecretsNone foundNo keys in source

Five audit layers

Gates applied: no behavioural pass, no license.

.claude/skills

Audit 0.4.1 · source sha 81994516a430 · 14 Oct 2026Report an issue or request a re-scan

• PublisherWILLOSCAR GitHub userOwner of the repository this skill is read from
• Source folderWILLOSCAR/research-units-pipeline-skills/.codex/skills/deliverable-selfloopThe exact folder we audited
• LicenceNot declaredAs the repository states it
• Commit pinned for this verdict81994516a430A new audit runs when the source changes
• Repository statusNot archivedLast push 9 days ago
• Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

Its description: Score an existing reader-facing deliverable and emit a deterministic PASS/FAIL scorecard plus bounded repair plan; use after synthesis, not to rewrite content.

---
name: deliverable-selfloop
description: Score an existing reader-facing deliverable and emit a deterministic PASS/FAIL scorecard plus bounded repair plan; use after synthesis, not to rewrite content.
---

# Deliverable Self-Loop

Runs the final quality gate for a reader-facing deliverable and always writes a PASS/FAIL report.

## Inputs

Primary input depends on the active pipeline contract:
- brief deliverable
- paper-review deliverable
- evidence-review deliverable
- idea memo bundle
- tutorial deliverable

## Output

- `output/DELIVERABLE_SELFLOOP_TODO.md`
- for `research-brief`: `output/BRIEF_SCORECARD.md` and `output/BRIEF_SCORECARD.json`
- for `paper-review`: `output/REVIEW_SCORECARD.md` and `output/REVIEW_SCORECARD.json`
- for `idea-brainstorm`: `output/IDEA_SCORECARD.md` and `output/IDEA_SCORECARD.json`
- for `evidence-review`: `output/EVIDENCE_SCORECARD.md` and `output/EVIDENCE_SCORECARD.json`

## Dispatch rule

The gate should dispatch by pipeline contract first:
- `quality_contract.deliverable_kind`

Only fall back to legacy profile-name checks when contract metadata is missing.

## Script boundary

`scripts/run.py` should:
- detect the active deliverable contract
- run the matching evaluator
- always write a report

It should not mutate the deliverable itself.

## Acceptance

- report exists
- report contains `- Status: PASS` or `- Status: FAIL`
- PASS only when the active deliverable satisfies its minimum section / artifact contract
- `research-brief` additionally requires valid core-set pointers and the configured score threshold
- `paper-review` additionally requires the configured score threshold and all critical rubric dimensions
- `idea-brainstorm` additionally requires traceable anchors, actionable lead directions, and the configured score threshold
- `evidence-review` additionally requires clause-linked screening, complete extraction rows, synthesis pointers, and the configured score threshold

## Non-goals

- rewriting the deliverable
- choosing upstream fixes beyond pointing to the missing contract items

Excerpt, first 2,065 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

git clone https://github.com/WILLOSCAR/research-units-pipeline-skills
git -C research-units-pipeline-skills checkout 81994516a430

Works in

1. 2026-10-14
   ✓SAFE B 89 / 100

   first audit · source 81994516a430

Questions about Deliverable Selfloop

• github.com/WILLOSCAR/research-units-pipeline-skills/tree/main/.c — the audited source at commit 81994516a430 and every finding aboveObserved · 14 Oct 2026

Spotted something wrong? Report a correction · Methodology