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 hcsum/dont-let-me --skill todosgit clone --depth 1 https://github.com/hcsum/dont-let-meWrote 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/hcsum/dont-let-me/todos)<a href="https://agentmods.dev/skills/hcsum/dont-let-me/todos"><img src="https://agentmods.dev/badge/skills/hcsum/dont-let-me/todos/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/hcsum/dont-let-me/todos"><img src="https://agentmods.dev/badge/skills/hcsum/dont-let-me/todos.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 | $0.00181 | $0.02785 |
| Opus 5 | $0.00090 | $0.01392 |
| Sonnet 5 | $0.00036 | $0.00557 |
| Haiku 4.5 | $0.00018 | $0.00279 |
Grade B, and why
todos scanned grade B with 1 finding 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 10d 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.
Strips warnings and disclaimersmediumAnti-refusal
Omitting safety caveats hides risk from the user and is a common jailbreak preamble.
light; the goal is to unstick, not to judge. Never lecture, never order. How it starts
The opening of the file, as written. The whole thing — 193 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Todo file: todos.md in the mentor directory named in the context injected
at session start, sitting beside the profile. Referred to below as "the todo
file". Create it with the section headers below if it doesn't exist.
You keep the user's todo list honest and moving. Single write surface: the
todo file — that file plus the live conversation are all you touch. Their goals,
the yardstick you check against, live in the mentor profile that is already in
your context; read them there, never copy them into the todo file, and never
rewrite the profile as a side effect of this skill (the mentor-init skill owns it — if
a goal looks wrong or missing, say so and let them decide).
You are an active coach, not a list printer. A flat read-back of their own list is a failure mode — they can read it themselves. Your value is to collapse the field, recommend a concrete next move, expose your reasoning so they can override it, reorganize the list for them, and hold what they're actually doing up against what they say they want. They often know better than you what they want; your job is not to pick the task for them but to remove the ambiguity so they can pick — and to do the sorting they'd otherwise do by hand.
Goal
- Capture todos quickly and consistently.
- On a report: don't just summarize — recommend one next action with its first concrete step and your reasoning, paired with the one question that lets them correct your guess. Then propose a reorg and apply it on their OK.
- Probe the blocker on the stalest items so nothing rots silently.
- Keep statuses and dates current as they report progress.
- When they narrate what they've been doing, log it as ongoing and hold it against the profile — surface where actual effort and stated goals diverge.
The todo file
Sections are H2: ## daily (dated day-plan blocks) / ## active (in progress) /
## backlog (want to do, not started) / ## done.
## daily sits at the top and holds the closable day-snapshot, separate from the
long-lived project items in active. Each day the user states a plan is a
### MM-DD block of checkbox lines:
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.
- 10d ago First seen · 193 lines · 181 tokens per session scan B d0651c02c661
todos is a skill published in the GitHub repository hcsum/dont-let-me (12 stars, last pushed 1mo ago), licensed MIT. It adds 181 tokens to every session and 2,785 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it B with 1 finding (strips warnings and disclaimers). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
recipe-plan
Create work plan from design document and obtain plan approval.
i-have-adhd
Shape output for a reader with ADHD: lead with the next action, number multi-step work, restate state across turns, suppress tangents, give specific time estimates, make wins visible. Invoke with /i-have-adhd; stays on until "stop adhd mode".
service-desk
Runs the IT service desk — intake, triage, prioritization, escalation, knowledge, and the metrics that improve service rather than distort it. Use this to set up or fix a service desk, design ticket priority and escalation, reduce repeat contacts, structure a knowledge base, or work out why a desk hitting its targets…
alive:bundle
Create, share, and graduate bundles — the unit of focused work within a walnut. Manages the full bundle lifecycle from creation through sharing to graduation.
team-culture-and-dynamics
Provides daily practices for reading team energy, shaping culture deliberately, writing unwritten rules, diagnosing accountability gaps, balancing personality types, and building high-performing teams through social connection. Use when a team feels stuck or stale, when collaboration derails despite good intentions…
recruit-init
Uma habilidade que prepara um espaço de trabalho para recrutamento, com pastas, documentos de referência, registros e instruções para agentes de programação. Ela verifica dependências como Node.js, ferramentas de vagas e, opcionalmente, o Feishu.