Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

/review-proposals — measured AI agent skill | DecimalAI

anthropics/review-proposals

NewSkillScoreMaintained

Review and approve (or reject) pending playbook update proposals from the playbook-monitor agent and apply approved changes to the practice profile. Use when the playbook-monitor agent has surfaced proposals, when the user says "review playbook proposals", "what playbook updates are pending", or wants to step through deviation-driven playbook changes.

agent-toolingby anthropics

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Legal & Contracts pack

decimalai skills pull anthropics-review-proposals

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/anthropics-review-proposals/SKILL.md

Raw file: /s/anthropics-review-proposals/SKILL.md ↗ · machine-readable summary at /s/anthropics-review-proposals.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 2h ago

—-33% tokens0% turns55% of whole cases pass with skill (12/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-05	✗→✓	▲ Improved	-64%	0%
case-10	✗→✓	▲ Improved	-4%	0%
case-18	✗→✓	▲ Improved	-50%	0%
case-22	✗→✓	▲ Improved	76%	0%
case-08	✓→✗	▼ Worse	-59%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

anthropics / claude-for-legal

commercial-legal/skills/review-proposals/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 8/3/2026Version v1

Skill contents

SKILL.md

/review-proposals

Steps through pending playbook update proposals from the monitor agent and applies approved changes to ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md.

Instructions

Load the playbook-monitor agent and run Step 5 (review and approval flow).

If no proposals file exists or it is empty: respond "No pending proposals. Playbook is up to date." Do not proceed further.

Present proposals one at a time. For each, show the full proposal block and offer four options: Accept, Reject, Edit, Defer.

For Accept or Edit: show the exact diff to ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md before writing. Only apply after the attorney explicitly confirms.

For Reject or Defer: log the decision. Do not modify ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md.

After all proposals are resolved: show a summary of what changed, then archive the proposals file.

Examples

/commercial-legal:review-proposals

/commercial-legal:review-proposals
(runs automatically after playbook-monitor notifies you)

Want to see if anthropics/review-proposals would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.