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 skills add thienhoc/dwi-by-thienhoc --skill dwi-budgetgit clone --depth 1 https://github.com/thienhoc/dwi-by-thienhocWrote 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/skills/thienhoc/dwi-by-thienhoc/dwi-budget)<a href="https://agentmods.dev/skills/thienhoc/dwi-by-thienhoc/dwi-budget"><img src="https://agentmods.dev/badge/skills/thienhoc/dwi-by-thienhoc/dwi-budget/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/skills/thienhoc/dwi-by-thienhoc/dwi-budget"><img src="https://agentmods.dev/badge/skills/thienhoc/dwi-by-thienhoc/dwi-budget.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 | $0.00056 | $0.00526 |
| Opus 5.5 | $0.00022 | $0.00210 |
| Sonnet 5.5 | $0.00011 | $0.00105 |
| Haiku 4.5 | $0.00006 | $0.00053 |
Grade A, and why
dwi-budget 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 1mo 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 — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dwi Budget
Make resource boundaries visible without turning them into performance theater.
Establish the boundary
Before substantial work, name the smallest useful budget the task needs:
- outcome boundary;
- context boundary;
- tool-call or stage boundary when relevant;
- time or checkpoint boundary when available;
- evidence required before spending more.
Do not ask for a numeric budget when a simple scope boundary is clearer.
Measurement labels
Report only fields supported by current evidence:
| Field | Meaning |
|---|---|
| Gross input | All input tokens reported by the harness |
| Cached input | Input tokens explicitly reported as cache reads |
| New input | Gross input minus cached input, only when both are comparable |
| Output | Output tokens reported by the harness |
| Wall time | Elapsed time measured for the stated scope |
| Tool calls | Calls counted in the stated stage or task |
| Coordination | Time, calls, or stages spent handing work across agents |
If a field is unavailable, mark it UNKNOWN or UNAVAILABLE. Do not estimate it silently.
Cache and cost rules
- Cached input is reuse evidence, not automatically a monetary saving.
- A percentage needs a denominator and scope.
- Token counts from different providers or tools may not be comparable.
- Price changes over time; use a dated price source before calculating money.
- Do not convert one canary into a universal claim.
Spending protocol
- Use the cheapest evidence that can answer the current question.
- Reuse already-read context instead of repeatedly loading it.
- Prefer bounded queries over broad repository scans.
- Add a second agent only when independent perspective or disjoint work has a defined value.
- Checkpoint before crossing the stated boundary.
- Ask for more budget only with the new uncertainty and expected value.
Report format
Use:
Outcome: <what was achieved>
Boundary: <scope or budget used>
Observed: <supported measurements>
Unknown: <unavailable measurements>
Next spend: <none, or one justified step>
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 70 lines · 56 tokens per session scan A 5a4b29b19cbd
dwi-budget is a skill published in the GitHub repository thienhoc/dwi-by-thienhoc (3 stars, last pushed 17d ago), licensed Apache-2.0. It adds 56 tokens to every session and 526 once invoked, about $0.0002 per session on Opus 5.5. 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 skills, from other repositories
handoff
Compress project context into HANDOFF.md with workflow state, analysis checkpoint, next gate, verification, current hypothesis, freshness anchors, optional gotcha references, and a resume packet for the next agent session.
operator-profile
Create a compact operator preference profile that tunes safe workflow defaults without overriding task contracts, repository instructions, safety boundaries, or required verification.
eco-mode
Keep a small, budgeted working context for low-cost model execution without weakening task contracts, required sources, or verification.
ask-matt
Ask which skill or flow fits your situation. A router over the user-invoked skills in this repo.
create-prompt
Create prompts for sidecar workspaces. Covers prompt structure (name, ticketMode, body), template variables (ticket with fallbacks), config file locations (global vs project), and scope overrides. Use when creating or modifying prompts in sidecar config files.
usage-audit
Audit a Claude Code setup for token waste and context bloat. Checks MCP servers, CLAUDE.md, skills, and settings against bloat filters. Triggers on: "audit my context", "usage audit", "token audit", "context bloat". NOT for codebase audits.