commitments

commitments is a skill for Claude Code, Codex from fledgeling-co/fledgeling-plugins. It costs 50 tokens per session (600 once invoked), scanned A, original, MIT.

A workflow for finding promises and handoffs in selected conversations or documents. It distinguishes suggestions, accepted commitments, cancellations, owners, deadlines, dependencies, and evidence of completion.

In plain words
What is it for?
Use it to answer what was promised, who owns the next step, what another person is waiting for, and which commitments are still open.
Why use it?
It prevents an unanswered request or model-generated suggestion from being mistaken for a person's promise. It also preserves later changes instead of silently merging separate commitments.

Skill for Claude CodeCodex

Written for Claude Code and Codex: shipped in a Claude Code plugin, but also agents/openai.yaml present.

Part of the follow-through plugin — 5 skills shipped together

Good fit Use it to answer what was promised, who owns the next step, what another person is waiting for, and which commitments are still open.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/fledgeling-co/fledgeling-plugins/commitments
Install

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.

Any agent
npx skills add fledgeling-co/fledgeling-plugins --skill commitments
Clone the repo
git clone --depth 1 https://github.com/fledgeling-co/fledgeling-plugins

Made for: Claude Code, Codex.

Or install follow-through, the plugin that ships this one along with the rest of its 5 skills.

Wrote 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.

agentmods badge for commitments

README.md
[![agentmods](https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/commitments/github.svg)](https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/commitments)
Your own site
<a href="https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/commitments"><img src="https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/commitments/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.

agentmods 80×15 button for commitments

Your own site · 80×15
<a href="https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/commitments"><img src="https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/commitments.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 50 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 600 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00050 $0.00600
Opus 5.5 $0.00020 $0.00240
Sonnet 5.5 $0.00010 $0.00120
Haiku 4.5 $0.00005 $0.00060

Measured 5d ago against content hash fab5c4bb4b67, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

commitments 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 5d 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.

plugins/follow-through/skills/commitments/SKILL.md · 50 lines

What it actually says

Commitments and handoffs

Read the evidence contract. Freeze the selected people/conversations, dates and requested outcome. A review of commitments is read-only; an instruction to carry out a named commitment permits its bounded work. It does not grant authority to contact every person mentioned in historical activity.

Read enough of each selected conversation to establish the promise in context, including later amendments and cancellations. Use a connector/document source over history whenever available. Capture stable thread/message or document revision identity, speaker, promised deliverable, owner, recipient/dependency, due date with timezone, condition and acceptance evidence. Attribute statements to the speaker; do not turn a suggestion, unanswered request, model proposal or calendar event into a human promise. A relative date without a resolved timezone is a missing fact rather than a precise deadline.

Join repeated references to the same deliverable without merging two colleagues' independent promises. Preserve changes: a later agreement can supersede a date, but silence does not accept a handoff or cancel a promise. An outbound message proves attempted delivery, a remote acknowledgement proves acceptance, and the agreed artifact/effect proves completion. Recheck the current destination and artifact access before proposing a send; a display name alone is insufficient.

Give a short table of confirmed, proposed, waiting, completed and unverified commitments, ordered by deadline/dependency. Each row names who owes what to whom, the evidence, what changed and the next action. Include cancelled/superseded items when needed to prevent an old promise being revived. Do not declare an item overdue when its date or current status is unknown.

Prepare a local handoff packet with task, owner, input artifacts, scope, writable paths, tools, required output, acceptance evidence and stopping condition. A colleague-facing draft states the actual progress and precise blocker without promising work the user did not authorise. Sending email/chat, inviting people, assigning tasks or changing a tracker needs applicable authority. Existing agent session handoffs route to flagship:fleet-console; substantive feature readiness routes to shipyard:triage. Check the receiving runtime exposes the skill before claiming it ran.

Stop when the requested commitments are reconciled and authorised handoffs or deliverables have their evidence. If a request includes future reminders, use supported scheduling and deduplicate existing tasks; never silently start an ongoing colleague monitor. For retained state use the ledger format with domain commitment; observed completion must match the original deliverable.

Files

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.

Changes

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.

  1. 5d ago First seen · 50 lines · 50 tokens per session scan A fab5c4bb4b67

Subscribe to this mod's changes

commitments is a skill published in the GitHub repository fledgeling-co/fledgeling-plugins (2 stars, last pushed 4d ago), licensed MIT. It adds 50 tokens to every session and 600 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 2026-10-12.