Ready to supercharge your AI agents?

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

Get Started Free →

Findings Capture — measured AI agent skill | DecimalAI

a5c-ai/findings-capture

SkillScore36Maintained

Capture and persist research findings, discoveries, and decision rationale to findings.md.

by a5c-ai

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

Automatic

Knowledge Management pack

decimalai skills pull a5c-ai-findings-capture

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/a5c-ai-findings-capture/SKILL.md

Raw file: /s/a5c-ai-findings-capture/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-findings-capture.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 9d ago

—-30% tokens0% turns36% of whole cases pass with skill (8/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-04	✗→✓	▲ Improved	-38%	0%
case-06	✗→✓	▲ Improved	-24%	0%
case-12	✗→✓	▲ Improved	-75%	0%
case-13	✗→✓	▲ Improved	-63%	0%
case-14	✗→✓	▲ Improved	-57%	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 — 7/31/2026

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

a5c-ai / babysitter

library/methodologies/planning-with-files/skills/findings-capture/SKILL.md

View on GitHub ↗

License MITImported 7/31/2026Version v1

Skill contents

SKILL.md

Findings Capture

Capture and persist research findings, discoveries, and decision rationale to findings.md.

Agent

Findings Curator - pwf-findings-curator

Workflow

Initialize findings.md with structured section headers
Capture findings during execution with timestamps
Apply 2-Action Rule: persist after every 2 operations
Record decisions with supporting rationale
Deduplicate and organize on final flush
Analyze coverage for verification

Inputs

projectPath - Root path for planning files
phaseName - Current phase name
findings - Array of finding objects
decisions - Array of decision objects

Outputs

Updated findings.md with organized, timestamped entries
Coverage analysis for verification

Process Files

planning-orchestrator.js - Findings initialization and appending
planning-execution.js - 2-Action Rule batches and final flush
planning-verification.js - Coverage analysis

Want to see if a5c-ai/findings-capture 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.