Strands Agents is an open-source SDK for building and running AI agents in Python and TypeScript. Developers use it to create agents with model providers, tools, lifecycle controls, memory, sessions, streaming, tracing, and evaluations, and the catalogue includes add-ons for its agent-building workflow.
Borrowing it
Nothing to install: this file belongs to strands-agents/harness-sdk. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/strands-agents/harness-sdk/main/.agents/skills/pr-feedback/SKILL.mdgit clone --depth 1 https://github.com/strands-agents/harness-sdkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/strands-agents/harness-sdk/pr-feedback)<a href="https://agentmods.dev/skills/strands-agents/harness-sdk/pr-feedback"><img src="https://agentmods.dev/badge/skills/strands-agents/harness-sdk/pr-feedback/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/strands-agents/harness-sdk/pr-feedback"><img src="https://agentmods.dev/badge/skills/strands-agents/harness-sdk/pr-feedback.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00044 | $0.00618 |
| Opus 5.5 | $0.00018 | $0.00247 |
| Sonnet 5.5 | $0.00009 | $0.00124 |
| Haiku 4.5 | $0.00004 | $0.00062 |
Grade A, and why
pr-feedback scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PR Review Feedback
Fetch PR review feedback and inline comments, categorize them, and present options to fix.
Process
1. Determine the PR Number
Auto-detect from the current branch:
gh pr view --json number -q .number
If that fails (detached HEAD, no tracking branch), ask the user for the PR number or URL.
2. Fetch All Feedback
Use the bundled script to fetch reviews, inline comments, and issue-level comments in one shot:
bash .agents/skills/pr-feedback/fetch-pr-feedback.sh <number> [--repo owner/repo]
- Omit
<number>to auto-detect from the current branch. - Use
--repowhen the PR is in a different repo than the current directory.
The script returns JSON with three arrays:
reviews— top-level review summaries (non-empty bodies only)inline_comments— unresolved thread comments withupvotes/downvotesarrays (usernames who reacted),outdatedflag, anddiffHunkon the first comment in each threadcomments— issue-level comments
3. Summarize and Present to the User
Read all the feedback, use your judgment to group related items, and present a numbered list of things to address. Keep each item to one line. Put the most impactful items first.
Use upvotes, downvotes, and the PR author's own replies to determine priority:
- Upvoted by the PR author: This signals agreement — recommend fixing it.
- Author replied agreeing (e.g. "good point", "I'll fix", "makes sense"): Same as an upvote — recommend fixing it.
- Upvoted by other reviewers (not the author): Signals community agreement the issue matters — lean toward recommending.
- No signal from the author: Present as a suggestion but don't assume it should be fixed.
- Author replied disagreeing or explaining: Present for context but mark as "discussed — likely skip".
- Outdated comments: Group separately at the end — these may have been addressed by subsequent pushes.
When presenting the list, annotate items you recommend fixing (based on the signals above) so the user can quickly confirm "all recommended" or cherry-pick.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 62 lines · 44 tokens per session scan A ca1bef61ea88
pr-feedback is a skill published in the GitHub repository strands-agents/harness-sdk (8,996 stars, last pushed today), licensed Apache-2.0. It adds 44 tokens to every session and 618 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
returns-policy
Answer return, refund, and warranty questions for electronics. Use when the user mentions returns, refunds, RMAs, warranty coverage, damaged items, or opened packaging.
technical-troubleshooting
Provide setup, troubleshooting, and maintenance guidance. Use when the user reports a device that won't power on, connectivity issues, setup questions, overheating, or maintenance concerns.
code-review
Reviewing code for defects: diff-first reading, severity classification, evidence over inference, finding caps, and terminal verdicts. Use when reviewing a diff, PR, or patch, or auditing someone else's work.
responding-to-review
Answering review findings: verify each claim before implementing, no performative agreement, per-item fix/reject/defer with evidence, one batched remediation round. Use when responding to review or remediation rounds.
code-review-recent-changes
Review recent changes since a fixed point (commit, branch, tag, or merge-base) across three independent axes - Standards, Spec, and Maintainability - producing severity-ordered findings with an explicit verdict. Use when the user wants to review a branch, a PR, or recent committed changes.
code-review
Use when the user asks for a code review of a change, branch or PR.