Ready to supercharge your AI agents?

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

Decision Documentation · a5c-ai | DecimalAI

documentationby a5c-ai

Decision Documentation

a5c-ai/decision-documentation

Source a5c-ai/babysitter ↗License MIT

Create Architecture Decision Records (ADRs) documenting significant technical choices with context, options, consequences, and sequential numbering.

SkillScore82Maintained

Automatic

Doc Ops pack

Last verified 7/31/2026

What it changes

Passed 18 of 22 test cases with the skill, and 0 of 22 without it.

Without skill0%

With skill82%

Checks it passed that the model alone missed

Contains explicit Options and Consequences sections
Provides explicit rationale within a Decision section
Outputs to docs/decisions/ using sequential numbering

and 21 more in the full report

18 fixed0 passed either way4 still failing0 broke

Pass-rate lift: +82 pp
Tokens: -10%
Agent turns: 0%
Response time: -14%

Model gemini-3.6-flashMeasured 7/31/2026v118 of 22 cases passed with skill

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/a5c-ai-decision-documentation/SKILL.md https://app.decimal.ai/s/a5c-ai-decision-documentation/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add a5c-ai-decision-documentation, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/a5c-ai-decision-documentation/ and pins it in .decimal/skills.lock

decimalai skills add a5c-ai-decision-documentation

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/a5c-ai-decision-documentation/SKILL.md https://app.decimal.ai/s/a5c-ai-decision-documentation/SKILL.md

Raw file: /s/a5c-ai-decision-documentation/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-decision-documentation.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 18 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 18

Prompt

We just decided to switch our microservice communication from REST to gRPC to improve latency. Can you draft an architectural decision record capturing why we made this choice, the competing approaches we evaluated, and the expected tradeoffs?

What was checked, without and with the skill
Checked	Without skill	With skill
Specifies or creates the output file path under docs/decisions/ with a sequential numeric prefix	Missed	Passed
Includes headers for Title, Status, Context, Options, Decision, and Consequences	Missed	Passed
Sets Status to a valid state such as Proposed, Accepted, Deprecated, or Superseded	Passed	Passed

case-01 · gemini-3.6-flash · 7/31/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md20 linessingle file

Choosing between competing approaches
Documenting a decision for future reference
Recording rationale that would otherwise be lost

ADR Format

Title: Clear description of the decision
Status: Proposed, Accepted, Deprecated, Superseded
Context: Why this decision is needed
Options: Alternatives considered with pros/cons
Decision: The chosen option with rationale
Consequences: Expected positive and negative outcomes

Output

ADRs are written to docs/decisions/ with sequential numbering (e.g., 0001-use-typescript.md).

Tool Use

Invoke via babysitter process: methodologies/rpikit/rpikit-decision

Track record

Benchmark

18 / 22

test cases passed, vs 0 / 22 without

Adoption

Users: 1
Times used: 14

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/31/2026

View 15 scanner checks

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/rpikit/skills/decision-documentation/SKILL.md

View on GitHub ↗

License MITImported 7/31/2026Version v1

More skills from the same source and nearby in the registry, with their headline benchmark lift.