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.
npx agentmods add commands/vrennat/developersdevelopers/debtgit clone --depth 1 https://github.com/vrennat/developersDevelopersWhat 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 | $0.00054 | $0.00600 |
| Opus 5 | $0.00027 | $0.00300 |
| Sonnet 5 | $0.00011 | $0.00120 |
| Haiku 4.5 | $0.00005 | $0.00060 |
Grade A, and why
debt 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 yesterday.
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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/debt
Surface the deliberate routing shortcuts the workflow has taken. A // dD: marker (# dD: for Python/shell) records a borderline classification call /impl made on purpose — routing simple where medium was plausible, treating stakes as normal where high was arguable, skipping a review pass on a borderline change. Each marker should name an upgrade trigger: the condition under which the shortcut stops being safe. /debt collects them so the debt stays visible instead of silently compounding.
Procedure
-
Search the repo for markers (ripgrep preferred; fall back to grep):
rg -n -- 'dD:' . # or: grep -rn 'dD:' .Match both
// dD:and# dD:forms. -
Drop matches that document the convention rather than incur debt: this plugin's own files (
commands/,skills/,agents/,hooks/,README.md, and spec/plan docs that explaindD:). A real marker sits next to source code it is justifying. -
For each surviving match, extract: file path, line number, and the marker text after
dD:. -
Classify each marker. It has an upgrade trigger if its text names a revisit condition — a clause using words like
if,when,once,until,escalate,upgrade,beyond,grows,past,exceeds, orover. If none is present, tag it[no-trigger]. -
Print a ledger grouped by file (file order, lines ascending):
<file> L<line> — <marker text> L<line> — <marker text> [no-trigger] -
End with exactly one summary line:
<N> markers, <M> with no trigger.
Rules
- Read-only.
/debtreports; it never edits code or removes markers. - If zero markers survive step 2, print
No dD: markers found.and stop — no summary line. - Do not invent triggers. A decision with no written revisit condition is
[no-trigger]by design; that gap is the signal worth surfacing.
Example
src/hand/sort.ts
L42 — routed simple, escalate if sorting logic grows beyond this file
L88 — skipped adversarial pass, borderline stakes [no-trigger]
2 markers, 1 with no trigger.
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.
- yesterday First seen · 55 lines · 54 tokens per session scan A e5293640b2e7
debt is a command published in the GitHub repository vrennat/developersDevelopers (2 stars, last pushed 1mo ago), licensed MIT. It adds 54 tokens to every session and 600 once invoked, about $0.0003 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.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
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.