agent-decision-receipts

Mint a tamper-evident, post-quantum-signed receipt for a consequential agent action (deploy, delete, pay, grant-access, model decision) so it can be verified later from the certificate alone. Use when an autonomous agent takes a side-effecting action that may need to be proven later, or when satisfying EU AI Act Article 12 record-keeping. Three decisions: whether an action needs a receipt, minting it, verifying it. Signing is delegated to the open-source OpenAgentOntology package. Not after-the-fact log analysis; not a hosted notary; not a legal opinion.

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

—

The risk profile of this skill

Quality

Content

86%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, highly actionable skill body: executable Quick Start, a clear three-decision workflow with validation checkpoints, and a genuine one-level-deep reference split. Minor conciseness gains are available by trimming the redundant 'NOT' paragraphs and tightening the PQ rationale.

Suggestions

Collapse the three 'NOT log analysis / NOT a hosted notary / NOT a legal opinion' paragraphs (already in the description) into a single one-line boundary note to reclaim tokens.

Add an explicit verify-failure feedback loop ('if hash_ok or a leg fails, the action or signature was tampered with — re-mint from a trusted manifest before relying on it') to lift workflow_clarity to 5.

Trim the 'Why post-quantum, by default' prose in the body since the detailed rationale already lives in references/receipt-fields.md; link to it instead.

Dimension	Reasoning	Score
Conciseness	Largely lean with no basic-concept padding and copy-paste examples, but the three 'NOT log analysis / NOT a hosted notary / NOT a legal opinion' paragraphs duplicate the description and the post-quantum rationale runs a touch long.	4 / 5
Actionability	Quick Start gives copy-paste-ready pip install, a real build_action_manifest.py invocation with concrete flags, and mint/verify python -c one-liners; the required-key manifest table makes execution unambiguous.	5 / 5
Workflow Clarity	A clear three-decision sequence with a Decision-1 checklist and build-time validation (exit 2 on missing fields) plus a verify step; the destructive-operation cap does not bind since validation exists, but an explicit verify-failure recovery loop is only implicit.	4 / 5
Progressive Disclosure	The body is an overview that signals one-level-deep references clearly — references/receipt-fields.md (verified to exist) for the full schema and PQ rationale — and scripts/build_action_manifest.py (verified to exist) is invoked directly rather than inlined.	5 / 5
	Total	18 / 20 Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A precise, third-person description that concretely states what it does, when to use it, and how it differs from adjacent log/notary/legal skills. Trigger-term coverage is strong though a few natural synonyms are reserved for the body.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'deploy, delete, pay, grant-access, model decision' plus 'minting it, verifying it' and the need-receipt decision — giving comprehensive coverage of what the skill does.	5 / 5
Completeness	Explicitly answers both: 'what' ('Mint a tamper-evident, post-quantum-signed receipt...') and 'when' ('Use when an autonomous agent takes a side-effecting action that may need to be proven later, or when satisfying EU AI Act Article 12 record-keeping').	5 / 5
Trigger Term Quality	Strong natural action verbs (deploy, delete, pay, grant-access) appear, but synonyms present in the body (wire, refund, export, egress) are absent from the description, leaving a few natural terms missing.	4 / 5
Distinctiveness Conflict Risk	A clear niche (tamper-evident agent-action receipts) with explicit disambiguation — 'Not after-the-fact log analysis; not a hosted notary; not a legal opinion' — minimizing overlap with related skills.	5 / 5
	Total	19 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: alirezarezvani/claude-skills
Path: ra-qm-team/skills/agent-decision-receipts/SKILL.md
Commit: 19392f7

Reviewed: ~1 hour ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.