Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/skyfox675/agents-skillsWrote 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/commands/skyfox675/agents-skills/spelunking-init-spec)<a href="https://agentmods.dev/commands/skyfox675/agents-skills/spelunking-init-spec"><img src="https://agentmods.dev/badge/commands/skyfox675/agents-skills/spelunking-init-spec/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/commands/skyfox675/agents-skills/spelunking-init-spec"><img src="https://agentmods.dev/badge/commands/skyfox675/agents-skills/spelunking-init-spec.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00123 | $0.01035 |
| Opus 5 | $0.00062 | $0.00517 |
| Sonnet 5 | $0.00025 | $0.00207 |
| Haiku 4.5 | $0.00012 | $0.00103 |
Grade A, and why
spelunking-init-spec 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 9d 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 — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/spelunking-init-spec — map the codebase into living specs
Operate in caveman mode (load the
cavemanskill) — a deep-dive spawns many agents over a large repo. Keep references,file:line, frontmatter, and the_index.mdtable byte-exact; the spec prose is the exception (humanizer — see the spelunking-specs skill).
Arguments: $ARGUMENTS
Parse them as:
- Optional paths or domain names scoping the dive (no args = the whole repo). Use a path/domain to deep-dive just that area.
- Optional
model:<tier>/effort:<level>, anywhere — these pin the recon agents' tier (defaultsonnet;opusfor large or tangled codebases).
Example: /spelunking-init-spec (whole app) · /spelunking-init-spec src/payments billing (scoped)
The spelunking-specs skill is the protocol — the numeric reference scheme (domain.subdomain.instance, an address, not a version; git versions the specs), the spec-file anatomy, the auto-discovered taxonomy, the reality+standards+drift content, and the verification bar all come from it. Consult it before starting.
Steps
-
Locate the spec set. Find
<specs-dir>(defaultspecs/). If_index.mdexists, read it first — a deep-dive resumes and extends the existing set rather than overwriting it (numbers are never reassigned). If absent, start fresh. -
Map & propose the taxonomy. Spawn survey agent(s) to walk entry points, module/package boundaries, routes, schemas, and build config and propose the domain list auto-discovered from the code's own vocabulary — no seed list. Assign each accepted domain the next free number and seed
_index.md. Detect the stack while mapping: if behaviour is convention-, config-, or runtime-wired, the survey and authoring agents run the spelunking-specs skill's spelunk loop — enumerate machine-names via the framework's introspection CLI, follow them by string across source, config, and templates, and observe the live page (browser MCP + the framework's debug output) — instead of static call-graph reading. Allowlist whatever introspection CLI the stack ships in.claude/settings.json(a routes/container/config-debug command, a module/package list) plus the browser MCP; with no running instance or CLI access, recon degrades to static + committed config and marks runtime-only claimsdraft.
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.
- 9d ago First seen · 35 lines · 123 tokens per session scan A 556e01ce1eb4
spelunking-init-spec is a command published in the GitHub repository skyfox675/agents-skills (10 stars, last pushed 8d ago), licensed MIT. It adds 123 tokens to every session and 1,035 once invoked, about $0.0006 per session on Opus 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 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.