agentleFS
ractiveractive/hyalo/.claude/skills/dogfood/SKILL.md

dogfood

Skill29 starsChanged 2 months agoMITView on GitHub

Run a dogfooding session for hyalo — build from source, exercise the CLI against real knowledgebases (own KB, MDN, GitHub Docs, VS Code docs), find bugs, verify recent fixes, assess UX, measure performance, and write a structured report. Use this skill whenever the user says /dogfood, "dogfood", "run a dogfood session", "test hyalo against real data", or wants to exercise hyalo on real knowledgebases to find issues.

  • Installs packages

81 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use dep-refs, feature-boundary-planner and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • axiom-desuSkill

    dep-refs

    Use when needs to inspect TypeScript export references in the z-code workspace, list exports from a file, verify whether an export is unused before deletion, investigate who imports a symbol during refactors, or combine pnpm knip unused-export results with pnpm dep:refs symbol-level reference tracing.

    Also in 6 projects that use this

  • axiom-desuSkill

    feature-boundary-planner

    Map a ZCode behavior change to current UI surfaces, state owners, protocol commands, persistence, and validation. Use for impact analysis, product-boundary planning, or an implementation handoff grounded in the checked-out source.

    Also in 6 projects that use this

  • lir-navcooSkill

    yuanbao

    Yuanbao (元宝) groups: @mention users, query info/members.

    Also in 18 projects that use this

  • Cartograf666Skill

    architecture-governance

    Bootstrap, document, enforce, and review a project's architecture through layer responsibilities, dependency direction, architecture manifests, legacy baselines, Change Envelopes, cross-layer approval policy, and deterministic architecture checks. Use when starting a project, auditing an existing codebase, planning cross-layer work, preventing architecture erosion, reviewing structural changes, or deciding whether implementation may expand into another layer.

    Also in 6 projects that use this

  • math-incSkill

    minecraft-modpack-server

    Set up a modded Minecraft server from a CurseForge/Modrinth server pack zip. Covers NeoForge/Forge install, Java version, JVM tuning, firewall, LAN config, backups, and launch scripts.

    Also in 5 projects that use this

  • math-incSkill

    pokemon-player

    Play Pokemon games autonomously via headless emulation. Starts a game server, reads structured game state from RAM, makes strategic decisions, and sends button inputs — all from the terminal.

    Also in 5 projects that use this

  • +1 more from math-inc

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 6 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 81 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read it to learn: which version was tested, which bugs were found, what was verified.

    line 40
  2. Very specific:

    Discover what changed since then: Use git log --oneline from that report's date to HEAD.

    line 41
  3. Distinctive:

    Read the most recent reports and collect any bugs marked as found (not yet verified fixed).

    line 53
  4. Distinctive:

    Pick target knowledgebases: Use a mix of:

    line 54
  5. Distinctive:

    Create a snapshot index for large vaults (500+ files).

    line 61
  6. Distinctive:

    Test the features you discovered in Phase 1 from the iteration files and git log.

    line 69
  7. Distinctive:

    Do help texts match actual behavior?

    line 97
  8. Very specific:

    Use hyalo set to ensure the frontmatter is correct.

    line 165
  9. Very specific:

    Check data quality: Use lint, find --broken-links, find --orphan to surface issues in the KB itself — this is valuable dogfooding too.

    line 179

“Common” is measured: kept by at least 6 in 10 of the 81 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Dogfood Session

A dogfood session exercises hyalo against real knowledgebases to find bugs, verify fixes, assess UX, and discover improvement opportunities. Sessions should feel like a real user working with the tool — not a mechanical checklist. Be curious, follow interesting leads, try edge cases that occur to you naturally.

Phase 1: Prepare

  1. Build and install from source, then prove the hyalo on PATH is this tree. The PATH binary is a cargo install and goes stale silently: on 2026-09-04 it was three days and eleven PRs behind target/release/hyalo, and a DEC-284 check against it nearly got filed as a bug.
    cargo build --release
    cargo install --path crates/hyalo-cli --target-dir target --force --locked
    test "$(hyalo --version | awk '{print $3}' | tr -d '(' | sed 's/+dirty$//')" = "$(git rev-parse --short=12 HEAD)" \
      && echo "hyalo on PATH is HEAD" || { echo "STALE hyalo on PATH: $(which hyalo)"; exit 1; }
    --target-dir target reuses the release build so the install takes seconds; the version string is hyalo X.Y.Z (<12-char hash>[+dirty] <date>), and +dirty is fine — it only means uncommitted edits such as the report you are about to write. If the check fails, fix the PATH before doing anything else — every later command in this skill, and every explorer agent you spawn, must run the same binary. Never fall back to ./target/release/hyalo for some commands and hyalo for others in one session.
  2. Find the last dogfood report to establish the baseline:
    hyalo find --tag dogfooding --sort date --reverse --limit 1 --format text
    Read it to learn: which version was tested, which bugs were found, what was verified.
  3. Discover what changed since then: Use git log --oneline from that report's date to HEAD. Look for merged iteration branches (iter-N/...). For each, read the iteration file in the knowledgebase to understand what was implemented:
    hyalo find --tag iteration --property "date>=YYYY-MM-DD" --sort date --format text
    Then hyalo read <iteration-file> --format text for each to understand the features, bug fixes, and UX changes. These are your priority test targets.
  4. Check prior dogfood reports for open bugs to re-verify:
    hyalo find --tag dogfooding --sort date --reverse --limit 5 --format text
    Read the most recent reports and collect any bugs marked as found (not yet verified fixed).
  5. Pick target knowledgebases: Use a mix of:

    • Own KB (hyalo-knowledgebase/, ~250 files) — well-structured, has schemas/views
    • MDN Web Docs (../mdn/files/en-us/, ~14K files) — large, stress-tests performance
    • GitHub Docs (../docs/content/, ~3.5K files) — complex nested YAML frontmatter
    • VS Code Docs (../vscode/, ~1K files) — mid-size, different structure

    For external KBs, check they exist first (ls ../mdn/files/en-us/ > /dev/null 2>&1). Create a snapshot index for large vaults (500+ files).

Phase 2: Exercise the tool

This is the creative part. Don't follow a rigid script — think about what a real user would do with hyalo and try it. That said, make sure every session covers these categories:

New feature verification

Test the features you discovered in Phase 1 from the iteration files and git log. For each new feature or changed command, don't just check the happy path — try edge cases, wrong inputs, combinations with other flags. The iteration files describe what was implemented and often include acceptance criteria that suggest good test scenarios.

Bug regression testing

Re-verify bugs from the most recent dogfood report(s). Use the exact commands or scenarios that originally exposed them. Mark each as STILL FIXED, REGRESSED, or PARTIALLY FIXED.

Creative exploration

This is where you go beyond verification. Try things like:

  • Unusual filter combinations (--property + --tag + BM25 + --section)
  • Edge-case inputs (empty strings, special characters, very long queries)
  • Commands on files with unusual frontmatter (nested objects, empty values, unicode)
  • --jq queries to build ad-hoc reports or dashboards
  • Views that chain with additional CLI flags
  • mv on files with many inbound links
  • lint --fix --dry-run on messy external KBs
  • Schema validation on write (--validate or validate_on_write)
  • Performance comparisons (indexed vs non-indexed, small vs large vault)
  • links fix on KBs with broken links
  • Anything that occurs to you as you go — follow your curiosity

UX assessment

Pay attention to:

  • Are error messages helpful? Do they suggest what to do?
  • Are hints useful and relevant?
  • Is --format text output scannable?
  • Do help texts match actual behavior?
  • Any friction points or confusing behavior?

Performance spot-checks

Time a few representative commands on each KB used:

time hyalo find --limit 1                    # baseline parse speed
time hyalo find "some query"                 # BM25 search
time hyalo summary                           # full vault scan
time hyalo find --property status=completed  # structured filter

Compare with prior reports if available. Flag regressions > 2x.

Phase 3: Write the report

Save to hyalo-knowledgebase/dogfood-results/ with this naming and frontmatter:

---
title: "Dogfood vX.Y.Z — <descriptive subtitle>"
type: research
date: YYYY-MM-DD
status: active
tags: [dogfooding, ...]
related:
  - "[[dogfood-results/previous-report]]"
  - "[[iterations/relevant-iteration]]"
---

Report structure

Organize findings into clear sections. Use this general shape, but adapt as needed:

# Dogfood vX.Y.Z — <subtitle>

Brief context: binary version, which KBs tested, file counts.

## New Feature Verification
### Feature Name (iter-NNN) — WORKING / BROKEN / PARTIAL
Description of what was tested and results.

## Bug Regression Testing
### BUG-N: Description — STILL FIXED / REGRESSED
Commands tested, results observed.

## Bugs Found
### BUG-N: Description (SEVERITY)
Steps to reproduce, expected vs actual, impact.

## UX Issues
### UX-N: Description (SEVERITY)
What's confusing or could be better.

## What Worked Well
Highlight things that are genuinely good — performance wins, helpful errors, etc.

## Performance
Table of timings if measured.

Severity levels: HIGH (data loss, wrong results), MEDIUM (confusing behavior, workaround exists), LOW (cosmetic, minor friction).

After writing the report

Use hyalo set to ensure the frontmatter is correct. Add related links to relevant iterations or prior dogfood reports.

Tips for good dogfood sessions

  • Follow the hints. Hyalo outputs drill-down suggestions — use them like a real user would.
  • Don't skip external KBs. Many bugs only surface at scale or with unusual frontmatter. The own KB is too well-structured to catch everything.
  • Note what's NOT there. Missing features, things you wished you could do — these are just as valuable as bugs.
  • Be specific. Include exact commands, exact error messages, exact output. Vague reports like "search felt slow" aren't actionable.
  • Test across KBs. A command that works on the own KB might fail on MDN due to scale or on GitHub Docs due to nested YAML.
  • Check data quality. Use lint, find --broken-links, find --orphan to surface issues in the KB itself — this is valuable dogfooding too.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.