local-review
Use for Plannotator plan or local diff review, or for a short Pi tree exploration followed by returning to implementation.
Skill by sounkou-bioinfo
Chalk Labs
Need help deploying local-review?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/sounkou-bioinfo-local-review
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/local-review
What's inside
- Skill name
local-review- Description
- Use for Plannotator plan or local diff review, or for a short Pi tree exploration followed by returning to implementation. Keep publishing explicit.
- License
- MIT
- Instructions
- 53 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Explore, review locally, then publish deliberately Use upstream Plannotator in Pi and the native session tree. Operating details and Remote-SSH forwarding are in [the operating guide](../../docs/operations.md#local-review). ## Build shared understanding - Agree the objective and evidence needed for success on the implementation branch. - For a short side discussion, have the user label an assistant anchor in `/tree`. Explore read-only for a few exchanges: alternatives, evidence, and uncertainties. - Before returning, give a concise brief: objective; decisions actually made by the user; observations and their source paths; unresolved questions; proposed next step. Keep suggestions distinct from approved decisions. - The user returns through `/tree` and chooses whether to carry a branch summary. The old branch remains inspectable. The user then chooses the implementation step. - Tree navigation does not restore files or Git state or cancel jobs. Isolate experiments explicitly. Close or reconcile outstanding review tabs before moving. ## Review locally - Use `/plannotator-plan-mode` and `plannotator_submit_plan` when a plan needs review. - Use `/plannotator-review --git` for the local diff, `/plannotator-annotate <path>` for Markdown, and `/plannotator-last` for an explanation. Use Pi's integration; a separate Plannotator CLI is not required. - Keep small tasks lightweight; a short clarification does not require a formal plan. - Run local checks and inspect meaningful evidence before treating work as ready. Include untracked files and the intended comparison base in local diff review. - Treat code-review annotations as findings to verify and discuss before editing. Ground verdicts in the code, then apply the revisions the user chooses. Browser feedback is human input, not proof that a claim is correct. Ask about unresolved consequential questions instead of guessing. ## Publish only when requested Local plan approval is not permission to push, create PRs, submit stacks, post remote reviews or comments, trigger remote CI, or merge. Obtain an explicit user request for publication. A request to inspect an existing PR does not by itself authorize posting feedback. Until then, keep review and verification local, without coworker notifications or CI spending. When publication is requested, use GitHub's native stacked PRs for coherent, dependent changes. Each layer should explain its decision, checks, and remaining uncertainty. GitHub owns published review and merge status; Plannotator annotations are not GitHub approvals. Bound review debt rather than merely dividing it among PRs. ## Remote development Support local machines and VS Code Remote-SSH. Keep review listeners on loopback and forward the port privately. Preserve VS Code's `BROWSER` helper. Choose an unused, session-specific port if a stable URL is needed. Do not expose an unauthenticated review server publicly or alter the user's SSH configuration.
From sounkou-bioinfo/pi-skills, MIT.
More from sounkou-bioinfo/pi-skills
- 0.598score
architecture-refinement
sounkou-bioinfo
Use for consequential architecture review, exploratory coding and theory-building when requirements or behavior are uncertain, technical issue refinement, and…
SkillAFreshness: A★ 2Claude Code +9 - 0.563score
bioinformatics-rewrites
sounkou-bioinfo
Use when reusing, binding, porting, or redesigning a bioinformatics implementation - reuse decisions, native FFI/bindings, compatibility with a named upstream…
SkillBFreshness: B★ 2Claude Code +9 - 0.561score
duckqc-design
sounkou-bioinfo
Use when designing DuckDB-native sequencing QC metrics, reductions, or compatibility outputs.
SkillBFreshness: B★ 2Claude Code +9 - 0.561score
duckvep-design
sounkou-bioinfo
Use when designing DuckDB-native variant consequence prediction, including transcript and haplotype semantics.
SkillBFreshness: B★ 2Claude Code +9 - 0.561score
s7-development
sounkou-bioinfo
Use when implementing or changing R S7 classes, properties, methods, or package integration.
SkillBFreshness: B★ 2Claude Code +9 - 0.559score
duckdb-c-extension
sounkou-bioinfo
Use for C DuckDB extension architecture, ownership, concurrency, API compatibility, vendoring, function catalogs, or packaging and exposing the extension…
SkillBFreshness: B★ 2Claude Code +9 - 0.556score
r-package-development
sounkou-bioinfo
Use for R package metadata, documentation generation, builds, checks, or release work.
SkillBFreshness: B★ 2Claude Code +9 - 0.556score
r-c-anti-slop
sounkou-bioinfo
Use to audit R/C helper sprawl, redundant checks, complexity, or host-unsafe assertions with Tree-sitter.
SkillBFreshness: B★ 2Claude Code +9 - 0.556score
duckhts-development
sounkou-bioinfo
Use for DuckHTS implementation, build, or public SQL/R API changes.
SkillBFreshness: B★ 2Claude Code +9 - 0.556score
duckhts-wasm-debugging
sounkou-bioinfo
Use when debugging DuckHTS in webR, duckdb-wasm, or Emscripten/browser workers.
SkillBFreshness: B★ 2Claude Code +9 - 0.556score
biomedical-evidence-search
sounkou-bioinfo
Use for variant, gene, trait, eQTL/PheWAS, score, or literature lookups via biomedical_search.
SkillBFreshness: B★ 2Claude Code +9 - 0.554score
project-orientation
sounkou-bioinfo
Re-establish working context after a handoff, when resuming unfamiliar repository work, or when requirements, source, tests and remembered decisions conflict.
SkillBFreshness: B★ 2Claude Code +9
Signals
- Stars
- 2
- Forks
- 1
- Open issues
- 2
- Last commit (this folder)
- Sep 22, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/sounkou-bioinfo/pi-skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from sounkou-bioinfo
- 0.598score
architecture-refinement
sounkou-bioinfo
Use for consequential architecture review, exploratory coding and theory-building when requirements or behavior are uncertain, technical issue refinement, and…
SkillAFreshness: A★ 2Claude Code +9 - 0.563score
bioinformatics-rewrites
sounkou-bioinfo
Use when reusing, binding, porting, or redesigning a bioinformatics implementation - reuse decisions, native FFI/bindings, compatibility with a named upstream…
SkillBFreshness: B★ 2Claude Code +9 - 0.561score
s7-development
sounkou-bioinfo
Use when implementing or changing R S7 classes, properties, methods, or package integration.
SkillBFreshness: B★ 2Claude Code +9 - 0.561score
duckvep-design
sounkou-bioinfo
Use when designing DuckDB-native variant consequence prediction, including transcript and haplotype semantics.
SkillBFreshness: B★ 2Claude Code +9 - 0.561score
duckqc-design
sounkou-bioinfo
Use when designing DuckDB-native sequencing QC metrics, reductions, or compatibility outputs.
SkillBFreshness: B★ 2Claude Code +9 - 0.559score
duckdb-c-extension
sounkou-bioinfo
Use for C DuckDB extension architecture, ownership, concurrency, API compatibility, vendoring, function catalogs, or packaging and exposing the extension…
SkillBFreshness: B★ 2Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get local-review running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
local-review skill. Agent Mart. https://agentmart.dev/skills/sounkou-bioinfo-local-review (last updated 2026-10-24)