القائمة
Structured recommendation output using PROPOSAL blocks with evidence, change, risk, and status fields. Use for any deliverable-local task that surfaces actionable findings.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Read one deliverable folder and produce a structured 7-section content digest summarizing identity, scope, document kit, dependencies, references, semantic framework, and quality observations.
Transform raw concordance evidence atoms and risk inventory rows into typed publication concordance candidates for register freeze review.
Post-synthesis cross-section semantic concordance verification for rewritten DBM package outputs.
Review a human-prepared draft DBM against the governed knowledge base — build review substrate, prepare candidate findings for human disposition.
Post-authoring evidence bundle and review for pipeline-produced DBM sections — build review substrate, prepare candidate findings for human readiness judgment.
Assemble the full rewritten DBM package from approved section outputs and produce package QA artifacts.
| name | proposal-format |
| description | Structured recommendation output using PROPOSAL blocks with evidence, change, risk, and status fields. Use for any deliverable-local task that surfaces actionable findings. |
| compatibility | Chirality TASK with DELIVERABLE_TASK profile; reasoning-only (no deterministic tools). |
| metadata | {"chirality-skill-version":"1","chirality-task-profile":"DELIVERABLE_TASK"} |
Produce structured, evidence-grounded recommendations for deliverable-local work using the PROPOSAL: block format. This skill standardizes how findings are expressed so they are reviewable, actionable, and traceable.
The PROPOSAL: pattern is the canonical recommendation format for deliverable-local tasks. It separates evidence from judgment, makes risk explicit, and preserves human authority over acceptance.
TASK with TaskProfile: DELIVERABLE_TASKDeliverablePath via DELIVERABLE_TASKTasks — specific asks; if omitted, the skill produces a baseline scan (see Method: baseline scan)RuntimeOverrides.MaxProposals — soft cap on proposals generatedRuntimeOverrides.FocusDocs — restrict analysis to named production docsRuntimeOverrides.IncludeLensTags — whether to include Lens: tags (default: only when UseSemanticLensing: true)| Key | Meaning | Default |
|---|---|---|
MaxProposals | Soft cap on proposals returned | 10 |
FocusDocs | Limit analysis to named docs (e.g., Specification.md,Guidance.md) | all production docs |
IncludeLensTags | Add Lens: tags even when semantic lensing is not active | false |
ProposalDepth | summary (title + status only) or full (all fields) | full |
This is a reasoning-only skill. No deterministic tools are required.
When combined with other skills that use deterministic tools (e.g., deliverable-consistency), the tool output provides the evidence base; this skill provides the output structure.
Disallowed behavior:
DELIVERABLE_TASKNEEDS_HUMAN_RULINGEvery recommendation MUST be expressed as a PROPOSAL: block with these fields:
- PROPOSAL: <short title>
- Evidence: <_REFERENCES.md item(s) and/or Source: <Doc> §<Heading>>
- Change: <precise change — what to add, modify, or remove>
- Why: <what it improves — clarity, completeness, verification, consistency, etc.>
- Risk: <downstream impact, dependency impacts, or conflicts>
- Status: <PROPOSED | APPLIED | NEEDS_HUMAN_RULING>
Optional field (when semantic lensing is active or IncludeLensTags: true):
- Lens: <Matrix.Row.Column> (or Lens: UNKNOWN if _SEMANTIC.md is missing)
location TBD. Never cite _SEMANTIC_LENSING.md as evidence — it is a worklist, not authority.PROPOSED — default; the human has not yet decidedAPPLIED — only when ApplyEdits: true and the edit was madeNEEDS_HUMAN_RULING — when the proposal involves a genuine trade-off, contradiction, or scope decisionWhen practical, group proposals by issue type:
After all PROPOSAL: blocks, always include:
- MISSING: <items where files or references are absent>
- NEEDS_HUMAN_RULING: <items requiring human decision — contradictions, trade-offs, scope questions>
- DEPENDENCY_NOTES: <cross-deliverable interfaces, blockers, mismatches>
Each section may be none if empty.
When Tasks is omitted from the brief, produce a baseline assessment:
MaxProposals) — the highest-value recommendations based on a sweep of the production documentsTBD items — the most impactful unresolved markersThis baseline scan is the default behavior, not a separate mode. It is what the skill does when given no specific direction.
PROPOSAL: blocks (structured per the format above)MISSING: itemsNEEDS_HUMAN_RULING: itemsDEPENDENCY_NOTES: itemsApplyEdits: true)MEMORY.md through DELIVERABLE_TASK closeoutTBD — do not invent content to fill gapsNEEDS_HUMAN_RULING, not reconciled silentlylocation TBD)