Catalog
JuliusBrussee/cavecrew

When to delegate to `cavecrew-investigator` (locate code), `cavecrew-builder` (1-2 file edit) or `cavecrew-reviewer` (diff review) instead of working inline or using `Explore`. Their output is compressed, so main context lasts longer.

NewUpdated Nov 7, 2026

Cavecrew = three subagent presets that answer in ultracave voice (tool runs: one status line in, one out; nothing between routine calls). Same job as Anthropic defaults (Explore, edit-style agents, reviewer); difference is the tool-result they return is compressed, so main context shrinks per delegation.

When to use cavecrew vs alternatives

Task Use
"Where is X defined / what calls Y / list uses of Z" cavecrew-investigator
Same but you also want suggestions/architecture commentary Explore (vanilla)
Surgical edit, ≤2 files, scope obvious cavecrew-builder
New feature / 3+ files / cross-cutting refactor Main thread, or feature-dev:code-architect if installed
Review diff, branch, or file for bugs cavecrew-reviewer
Deep code review with rationale + alternatives Code Reviewer (vanilla) if installed, else main thread
One-line answer you already know Main thread, no subagent

Rule of thumb: if you'd want the subagent's output in 1/3 the tokens, pick cavecrew. If you'd want prose, pick vanilla.

Why this exists (the real win)

Subagent tool results get injected into main context verbatim. A vanilla Explore that returns 2k tokens of prose costs 2k tokens of main-context budget every time. The same finding from cavecrew-investigator comes back as a compressed path:line table instead. Across 20 delegations in one session that's the difference between context exhaustion and finishing the task.

Output contracts

What main thread can rely on per agent:

cavecrew-investigator

<Header>:
- path:line — `symbol` — short note
totals: <counts>.

Or No match. Always file-path-first, line-number-attached, backticked symbols. Safe to grep with path:\d+.

cavecrew-builder

<path:line-range> — <change ≤10 words>.
verified: <re-read OK | mismatch @ path:line>.

Or one of: too-big. / needs-confirm. / ambiguous. / regressed. (terminal first token).

cavecrew-reviewer

path:line: <emoji> <severity>: <problem>. <fix>.
totals: N🔴 N🟡 N🔵 N❓

Or No issues. Findings sorted file → line ascending.

Chaining patterns

Locate → fix → verify (most common):

  1. cavecrew-investigator returns site list.
  2. Main thread picks 1-2 sites, hands paths to cavecrew-builder.
  3. cavecrew-reviewer audits the diff.

Parallel scout (when investigation is broad): Spawn 2-3 cavecrew-investigator calls in one message (different angles: defs vs callers vs tests). Aggregate in main thread.

Single-shot edit (when site is already known): Skip investigator. Hand exact path:line to cavecrew-builder directly.

What NOT to do

  • Don't accept or reject a subagent's absence claim (No match., No issues., "section missing", "unchanged") without one targeted check of the relevant file, search scope, or diff. Check cited line numbers are within the file before acting.
  • Don't use cavecrew-builder when you don't already know the file. Spawn investigator first or main thread will eat tokens passing context.
  • Don't chain cavecrew-investigator → cavecrew-builder for a 5-file refactor. Builder will return too-big. and you'll have wasted a turn.
  • Don't ask cavecrew-reviewer for "general feedback" — it returns findings only, no architecture opinions. Use Code Reviewer for that when installed.
  • Don't expect prose. Cavecrew output is structured, sometimes terse to the point of cryptic. If a human will read it directly, paraphrase.

Auto-clarity (inherited)

Subagents drop caveman → normal English for security warnings, irreversible-action confirmations, and any output where fragment ambiguity could be misread. Resume caveman after.

Files2
2 files · 13.9 KB

Select a file to preview

Overall Score

82/100

Grade

B

Good

Grades are signals, not a certification. Always review a skill yourself before use.

Safety

88

Quality

85

Clarity

82

Completeness

75

Summary

Cavecrew is a decision guide skill that instructs the main agent thread when to delegate code tasks to three specialized subagent presets (investigator, builder, reviewer) instead of handling work inline. It documents output contracts, chaining patterns, and guardrails to help agents optimize context usage by receiving compressed tool results rather than verbose prose.

Detected Capabilities

decision guidancecontext budget managementsubagent delegationtool result contract definition

Trigger Keywords

Phrases that agents use to match this skill to user intent.

delegate code searchcompress agent outputcontext budget optimizationsurgical code editscode review delegationsubagent chainingcavecrew investigatorcavecrew builder

Referenced Domains

External domains referenced in skill content, detected by static analysis.

www.apache.org

Use Cases

  • Deciding when to delegate code location searches
  • Determining if a task fits surgical 1-2 file edits
  • Planning code review delegation strategy
  • Chaining investigation → fix → verification workflows
  • Optimizing context budget across multi-step tasks
  • Understanding subagent output formats for parsing results

Quality Notes

  • Clear decision matrix comparing cavecrew agents to vanilla alternatives (Explore, Code Reviewer, main thread)
  • Well-defined output contracts with structured format (path:line, emoji severity, counted findings) enabling safe parsing
  • Explicit guardrails section warning against misuse (e.g., Don't use builder without knowing the file, Don't expect prose)
  • Practical chaining patterns (Locate→fix→verify, parallel scout) with concrete examples
  • Auto-clarity provision for security warnings and irreversible actions shows security-aware design
  • Environment variable override mechanism documented for model customization per agent
  • Two supporting files (README.md, LICENSE) present and aligned with content
  • Apache 2.0 licensing clearly stated
Model: claude-haiku-4-5-20251001Analyzed: Nov 7, 2026

Reviews

Add this skill to your library to leave a review.

No reviews yet

Be the first to share your experience.

Version History

  1. v2.3

    Content updated

    ✦ AIAdds guidance against accepting subagent absence claims without verification.

    2026-11-07

    LATEST
  2. v2.2

    Content updated

    ✦ AISubagent results now returned as compressed path:line table instead of full prose to reduce context overhead.

    2026-10-08

    View This Version
  3. v2.1

    Content updated

    ✦ AIClarifies cavecrew output format to one-line tool status per call, and adds conditional install checks for vanilla alternatives in usage table.

    2026-10-03

    View This Version
  4. v2.0

    Contract changed: description

    ✦ AIClarified trigger conditions for delegation to cavecrew subagents; simplified compression rationale in description.

    triggering2026-09-09

    View This Version
  5. v1.0

    2026-08-17

    View This VersionInitial version

Use JuliusBrussee/cavecrew in your dev environment

Command Palette

Search for a command to run...