Ready to supercharge your AI agents?

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

Get Started Free →

Address PR Comments · fallow-rs | DecimalAI

code reviewby fallow-rs

Address PR Comments

fallow-rs/address-pr-comments

Source fallow-rs/fallow ↗License MIT

Triage and address GitHub PR review feedback for fallow, then implement the agreed fixes.

SkillScore61Maintained

Automatic

Coding pack

Last verified 9/28/2026

What it changes

Passed 10 of 17 test cases with the skill, and 9 of 17 without it.

Without skill52%

With skill61%

Checks it passed that the model alone missed

The response selects cargo test -p parser.
The comment is classified as already resolved.
The summary details which review comments were addressed.

and 4 more in the full report

6 fixed8 passed either way5 still failing4 broke

Pass-rate lift: +9 pp
Tokens: -34%
Agent turns: 0%
Response time: -22%

Model gemini-3.6-flashMeasured 9/28/2026v2 · latest is v414 of 23 cases passed with skill · 17 comparable

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

This benchmark measured an earlier version of the skill.

A newer attempt on gemini-3.6-flash on 10/4/2026 did not produce a publishable comparison (not comparable). The snapshot above is the previously measured run.

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/fallow-rs-address-pr-comments/SKILL.md https://app.decimal.ai/s/fallow-rs-address-pr-comments/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add fallow-rs-address-pr-comments, 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/fallow-rs-address-pr-comments/ and pins it in .decimal/skills.lock

decimalai skills add fallow-rs-address-pr-comments

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

curl -fsSL --create-dirs -o .claude/skills/fallow-rs-address-pr-comments/SKILL.md https://app.decimal.ai/s/fallow-rs-address-pr-comments/SKILL.md

Raw file: /s/fallow-rs-address-pr-comments/SKILL.md ↗ · machine-readable summary at /s/fallow-rs-address-pr-comments.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 6

Prompt

I updated crates/parser/src/lib.rs based on review feedback on PR #310. Should I execute the full workspace test suite cargo test --workspace or a specific package test cargo test -p parser?

What was checked, without and with the skill
Checked	Without skill	With skill
The response selects cargo test -p parser.	Missed	Passed

case-09 · gemini-3.6-flash · 9/28/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md25 linessingle file

Address PR Comments

Use GitHub review metadata plus the local checkout.

Workflow:

Fetch PR metadata and review comments. Confirm that the local branch is at the PR head SHA.
Check each comment against the code at the PR head before any change. Trace the call sites for a "this can fail" claim. Give each check an evidence level from docs/development/quality-gates.md.
Sort each comment into one group:
- Act on: the check shows the problem. Only these comments change code.
- Consider: valid, but it has a cost or changes scope. Ask the maintainer.
- Noted: true, but no change is necessary now.
- Dismissed: wrong, a preference, or missing context. Give one reason.
Implement the Act on fixes, one comment at a time.
Run the smallest relevant validation.
Draft one reply per thread for the maintainer to approve. Never post a reply or resolve a thread. Lead with the fix and its commit. Thank a contributor in one short sentence.
Summarize each group, with the reason for each Dismissed comment.

Follow repo conventions from AGENTS.md: no AI attribution, signed commits when committing.

Track record

Benchmark

10 / 17

test cases passed, vs 9 / 17 without

Adoption

Users: 1
Times used: 32

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 9/30/2026

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

fallow-rs / fallow

.agents/skills/address-pr-comments/SKILL.md

View on GitHub ↗

License MITImported 9/30/2026Version v4

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

Ready to supercharge your AI agents?

See it in action

What's inside

Address PR Comments

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills