Agent skill

Dakota Factory

by projectbluefin in projectbluefin/dakota

Maintain task-relevant Dakota guidance: documentation accuracy, official-source verification, and skill auditing.

MITAuto-check passedResearch & Science

Install Dakota Factory

skills CLI
$ npx skills add projectbluefin/dakota --skill dakota-factory -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install projectbluefin/dakota dakota-factory --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/projectbluefin/dakota.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/dakota-factory .claude/skills/dakota-factory && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
dakota-factory
GitHub stars
180
Token cost
~1.1k tokens
SKILL.md length
511 words
Files
1
Skills in repo
10
Repo updated
First seen
Licence
MIT

At a glance

Maintain task-relevant Dakota guidance: documentation accuracy, official-source verification, and skill auditing.

  • Works in 5 steps: Check Relevance and Authority: Identify… → Docs Are the Model → Audit Skill Usability → …
  • Auditing skills
  • SKILL.md covers When to Use, When NOT to Use, Core Process and Invariants, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Dakota Factory is an agent skill from projectbluefin/dakota. Maintain task-relevant Dakota guidance: documentation accuracy, official-source verification, and skill auditing. Use when auditing skills or correcting durable guidance revealed by a task.

Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Research & Science, covering Fact-checking and source verification. The repository describes itself as: The buildstream for making Bluefin. The licence is MIT.

When your agent uses it

  • Auditing skills
  • Correcting durable guidance revealed by a task

Example prompts

  • “/dakota-factory”

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Check Relevance and Authority: Identify durable errors or missing constraints relevant to the task. Correct existing guidance when edits…
  2. Docs Are the Model
  3. Audit Skill Usability
  4. Verify Official Documentation
  5. Keep Changes Scoped: Include relevant documentation corrections with the code diff for review. Do not add unrelated writebacks, create a…

What it can do on your machine

Read from SKILL.md and the folder at commit bd5d913. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Dakota Factory loads about 1.1k tokens when it runs. Until then it costs about 51 tokens; SKILL.md has 511 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~51
When it runs · the whole SKILL.md, loaded when a task matches
~1.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from projectbluefin/dakota at commit bd5d913, republished under its MIT licence (© projectbluefin). 511 words, ~1,093 tokens.

Download SKILL.mdSave it as .claude/skills/dakota-factory/SKILL.md (or your agent's skills folder).
name
dakota-factory
description
Maintain task-relevant Dakota guidance: documentation accuracy, official-source verification, and skill auditing. Use when auditing skills or correcting durable guidance revealed by a task.

Dakota Factory Model

Keep documentation aligned with executable configuration. Correct durable guidance relevant to the task, without requiring a documentation change for every session.

When to Use

  • Ending any implementation, debugging, or triage session that revealed non-obvious behavior or traps
  • Creating, auditing, or refactoring skills in .agents/skills/
  • Updating architectural docs in docs/ to reflect running truth
  • Verifying library or CLI syntax against current official documentation

When NOT to Use

  • Read-only queries without authorization to change documentation
  • Pure dependency version updates handled automatically by Renovate

Core Process

  1. Check Relevance and Authority: Identify durable errors or missing constraints relevant to the task. Correct existing guidance when edits are authorized; otherwise report the finding without writing files.
  2. Docs Are the Model:
    • Skill files are evergreen procedures, not historical logs, ledgers, or backlogs.
    • Do not record session dates (e.g. 2026-08-01: we found X). Extract the timeless rule.
    • Do not maintain running issue tables or resolved checklists in skills. Gaps belong in GitHub issues; resolved items belong in git history.
  3. Audit Skill Usability: Ensure the affected skill in .agents/skills/ contains:
    • Valid YAML frontmatter with name and a descriptive description; optional metadata should identify real sources, not unavailable tool requirements
    • ## When to Use
    • ## When NOT to Use
    • ## Core Process
    • ## Invariants / ## Rules
    • ## Common Rationalizations
    • ## Red Flags
    • ## Verification
  4. Verify Official Documentation:
    • Identify the external tool and the version used by Dakota.
    • Read its current official documentation directly, or use Context7 when available.
    • Verify the specific syntax or behavior against the pinned source when needed.
    • Cite the official URL in relevant guidance or optional metadata.verified-sources.
  5. Keep Changes Scoped: Include relevant documentation corrections with the code diff for review. Do not add unrelated writebacks, create a commit, or publish a PR without authorization.
Show full SKILL.md (223 more words)Show less

Invariants

  • Task Boundary: Documentation changes must be useful, relevant, and authorized. A read-only task can end with findings alone.
  • Evergreen Truth: Workflows, elements, the Justfile, and tests are authoritative truth. Prose that disagrees with executable configuration is stale and must be excised.
  • Zero Writing to ublue-os/*: Absolute prohibition. Ask a human to report upstream manually.
  • Commit Trailer: Use Assisted-by:, never Co-authored-by:.

Common Rationalizations

RationalizationReality
"Every task needs a documentation diff."Update guidance when the task changes or corrects it; do not manufacture unrelated work.
"Recording the incident date gives good context."Dated incident logs turn documentation into an unmaintained changelog. Document the timeless failure mode and prevention rule instead.
"I know how this CLI tool works from memory."Verify current syntax in official documentation. Context7 is an optional retrieval tool, not a prerequisite.

Red Flags

  • Sections titled ## Lessons Learned with dated entries (2026-07-30: ...)
  • Running issue tables with live issue numbers inside skill files
  • Skills lacking ## Red Flags or ## Verification sections
  • Contradictions between docs/ and executable workflow YAML or Justfile

Verification

  • Documentation changes are task-relevant and authorized
  • No dated session logs or resolved checkmarks added to evergreen docs
  • External tool syntax is verified against official sources and cited where relevant
  • Skill frontmatter parses, referenced paths exist, and commands match available tooling
  • All commits use conventional format with Assisted-by: trailer

References

© projectbluefin, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .agents/skills/dakota-factory of projectbluefin/dakota.

Open the folder on GitHubat commit bd5d913

Compare with similar skills

Dakota Factory next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Dakota Factory compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Dakota Factory this skillprojectbluefin/dakota180—~1.1kAutomated safety check: PassMIT
Perplexity Web Searchdavila7/claude-code-templates33k11 repos~3.5kAutomated safety check: NotesMIT
Citation Verification GuideGalaxy-Dawn/claude-scholar5.7k2 repos~1.9kAutomated safety check: PassMIT
Article Fact Checkerdigoal/blog8.6k—~939Automated safety check: PassGPL-2.0
Deep Research Agent TeamImbad0202/academic-research-skills51k—~13kAutomated safety check: PassCustom licence
Docs Grounding Verifiermicrosoft/apm4k—~1.9kAutomated safety check: PassMIT

Similar skills

  • Perplexity Web Search

    davila7/claude-code-templates

    Runs web-grounded searches through Perplexity's Sonar models over OpenRouter for current events, recent literature and cited facts beyond the model's training cutoff.

    33k GitHub starsUsed in 11 repos~3.5k tokens
    Research & ScienceAuto-check: notes
  • Citation Verification Guide

    Galaxy-Dawn/claude-scholar

    Reference guidance for checking every citation in academic writing against canonical sources such as DOI, arXiv, CrossRef and Semantic Scholar, to catch fake or wrong references.

    5.7k GitHub starsUsed in 2 repos~1.9k tokens
    Research & ScienceAuto-check passed
  • 三层审查模型,逐段逐句验证文章真伪、证据链与逻辑结构。Use when the user asks to fact-check, verify, audit, or evaluate the credibility of an article, essay, report, opinion piece, social-media post, or any written claim —…

    8.6k GitHub stars~939 tokensUpdated 2 days ago
    Research & ScienceAuto-check passed
  • Deep Research Agent Team

    Imbad0202/academic-research-skills

    Runs a 13-agent pipeline for rigorous academic research, from forming the question through systematic search, synthesis, bias checks and an APA 7.0 report.

    51k GitHub stars~13k tokensUpdated yesterday
    Research & ScienceAuto-check passed
  • Official

    A skill your agent uses to verify CLAIM-LEVEL grounding of a documentation page (or set of pages) against the source code.

    4k GitHub stars~1.9k tokensUpdated yesterday
    Research & ScienceAuto-check passed
  • Fact Checking

    bradygaster/squad

    Review and validate claims using counter-hypothesis testing.

    3.3k GitHub stars~503 tokensUpdated yesterday
    Research & ScienceAuto-check passed

More from projectbluefin/dakota

All 10 skills in this repo
  • Dakota Buildstream

    projectbluefin/dakota

    BuildStream elements, junctions, patches, dependency graphs, and build failures in Dakota.

    180 GitHub stars~1k tokensUpdated today
    Auto-check passed
  • Dakota Extensions

    projectbluefin/dakota

    Package, update, and configure GNOME Shell extensions, Quick Settings panels, and schemas in Dakota.

    180 GitHub stars~1.6k tokensUpdated today
    Auto-check passed
  • Dakota Packaging

    projectbluefin/dakota

    Add, remove, or update native software built from source in Dakota, including Go, Rust, Zig, C/Meson, and binary releases.

    180 GitHub stars~1.1k tokensUpdated today
    Auto-check passed
  • Dakota Ujust

    projectbluefin/dakota

    Author safe end-user ujust recipes in files/just-overrides/default.just, including quoting, gum, JSON, and public-post confirmation.

    180 GitHub stars~1.1k tokensUpdated today
    Auto-check: notes
  • Dakota Image

    projectbluefin/dakota

    OCI layer assembly, boot testing, installer boundaries, VM work, and local OTA verification for Dakota images.

    180 GitHub stars~929 tokensUpdated today
    Auto-check passed
  • Dakota Workstation

    projectbluefin/dakota

    Dakota host Homebrew integration and workstation-specific services.

    180 GitHub stars~937 tokensUpdated today
    Auto-check passed

Questions about Dakota Factory

What does Dakota Factory do?

Maintain task-relevant Dakota guidance: documentation accuracy, official-source verification, and skill auditing. Dakota Factory is an agent skill from projectbluefin/dakota. Maintain task-relevant Dakota guidance: documentation accuracy, official-source verification, and skill auditing.

When should I use Dakota Factory?

Dakota Factory fits situations like: auditing skills; correcting durable guidance revealed by a task.

How do I install Dakota Factory in Claude Code?

Run `npx skills add projectbluefin/dakota --skill dakota-factory -a claude-code`. Or copy the skill folder (.agents/skills/dakota-factory in projectbluefin/dakota) into .claude/skills/dakota-factory in your project. Claude Code loads it when a task matches its description.

How do I install Dakota Factory in Codex?

Run `npx skills add projectbluefin/dakota --skill dakota-factory -a codex`. Or copy the skill folder (.agents/skills/dakota-factory in projectbluefin/dakota) into .agents/skills/dakota-factory in your project. Codex loads it when a task matches its description.

Can I use Dakota Factory in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add projectbluefin/dakota --skill dakota-factory -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/dakota-factory, .gemini/skills/dakota-factory, .github/skills/dakota-factory and .opencode/skills/dakota-factory in your project.

What does Dakota Factory need to run?

SKILL.md names no scripts, command-line tools or credentials: Dakota Factory is instructions for the agent only.

Does Dakota Factory access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Dakota Factory safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Dakota Factory use?

Dakota Factory is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Dakota Factory use?

About 1.1k tokens (SKILL.md is roughly 4.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Dakota Factory?

Skills that share tags, products or a category with Dakota Factory: Perplexity Web Search (davila7/claude-code-templates, 33k stars), Citation Verification Guide (Galaxy-Dawn/claude-scholar, 5.7k stars), Article Fact Checker (digoal/blog, 8.6k stars) and Deep Research Agent Team (Imbad0202/academic-research-skills, 51k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Dakota Factory?

projectbluefin (a GitHub organization) maintains it in projectbluefin/dakota, which has 180 GitHub stars. The repository holds 10 skills in this directory. The repository was last updated on October 11, 2026.

Source: projectbluefin/dakota on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.