Name: agent-decision-receipts
Rating: 67.2 (1 reviews)
Author: alirezarezvani

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

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

The canonical home for this skill is agent-decision-receipts in alirezarezvani/claude-skills

Quality

Content

71%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-organized, actionable skill body that sequences its three decisions clearly and provides copy-paste commands. The main defect is that referenced bundle files (scripts/ and references/) are missing, which undermines progressive disclosure despite good in-body structure.

Suggestions

Ship the referenced bundle files (scripts/build_action_manifest.py and references/receipt-fields.md) so the signaled one-level-deep navigation actually resolves, or remove the references if the skill is meant to be standalone.

Add an explicit error-recovery loop for the build step (e.g. "if build_action_manifest.py exits 2, fix the missing key and re-run") to give the destructive/consequential workflow a full validate->fix->retry feedback loop.

Trim the "NOT log analysis / NOT a hosted notary / NOT a legal opinion" section in the body since it duplicates the description's negative boundaries, to improve token efficiency.

Dimension	Reasoning	Score
Conciseness	Mostly lean and efficient with the tight "Three decisions, nothing else" framing and no padding about basic concepts, though the "NOT log analysis / NOT a hosted notary / NOT a legal opinion" section repeats material from the description and could be trimmed.	4 / 5
Actionability	Provides concrete, executable commands (pip install, build_action_manifest.py with full flags and exit code, inline mint_receipt/verify_receipt one-liners) covering the common cases, with only minor terseness in the inline python -c mint/verify snippets.	4 / 5
Workflow Clarity	The three decisions are clearly sequenced (need a receipt? -> mint -> verify) with a Decision 1 test table and verification as an explicit checkpoint (hash_ok, sig_ok, exit 2 on missing keys), though an explicit build-failure retry loop is absent for these destructive/consequential operations.	4 / 5
Progressive Disclosure	The body is well structured (Overview, Quick Start, three Decisions, Anti-Patterns, Cross-References) with a one-level-deep pointer to references/receipt-fields.md, but the referenced bundle files (scripts/build_action_manifest.py and references/receipt-fields.md) do not exist on disk, breaking the signaled navigation.	3 / 5
	Total	15 / 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 strong, specific description with explicit what/when guidance and clear negative boundaries that carve out a distinct niche. The main weakness is that several natural trigger phrases are wrapped in specialized cryptography jargon rather than plain user language.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ("deploy, delete, pay, grant-access, model decision") plus the mechanism (tamper-evident, post-quantum-signed receipt verified from the certificate alone) and the three decisions, giving comprehensive coverage.	5 / 5
Completeness	Explicitly answers both what ("Mint a tamper-evident...receipt...verified later from the certificate alone") 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") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Strong natural triggers ("Use when an autonomous agent takes a side-effecting action that may need to be proven later", "EU AI Act Article 12 record-keeping", deploy/delete/pay/grant-access) with good keyword coverage, though some specialized phrasing like "post-quantum-signed" is jargon a user may not naturally say.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (signed receipts for consequential agent actions) reinforced by explicit negative boundaries ("Not after-the-fact log analysis; not a hosted notary; not a legal opinion"), minimizing conflict with other skills.	5 / 5
	Total	19 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 1 missing	Warning
referenced_paths_exist	Referenced path issues: 3 missing	Warning

	Total	14 / 16 Passed

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

Reviewed: ~2 hours 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.