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 skills/chrislamdev/hermes-core-skills/context-aware-task-decompositionnpx skills add ChrisLamDev/hermes-core-skills --skill context-aware-task-decompositiongit clone --depth 1 https://github.com/ChrisLamDev/hermes-core-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/skills/chrislamdev/hermes-core-skills/context-aware-task-decomposition)<a href="https://agentmods.dev/skills/chrislamdev/hermes-core-skills/context-aware-task-decomposition"><img src="https://agentmods.dev/badge/skills/chrislamdev/hermes-core-skills/context-aware-task-decomposition.svg" alt="Measured on agentmods" 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 | $0.00020 | $0.00345 |
| Opus 5 | $0.00010 | $0.00172 |
| Sonnet 5 | $0.00004 | $0.00069 |
| Haiku 4.5 | $0.00002 | $0.00034 |
Grade A, and why
context-aware-task-decomposition 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.
What it actually says
Context-Aware Task Decomposition
Self-regulate task size based on context window capacity. When generating multi-step work, estimate if it fits in the remaining context and decompose accordingly.
Decision Flow
Is the task > 10 steps or likely > 2000 lines?
├─ Yes: Decompose into sub-tasks, dispatch via delegate_task()
│ Each sub-task gets its own full context window
└─ No: Execute directly
When to Decompose
- Complex code generation (multi-file)
- Long research with many sources
- Refactoring across multiple modules
- Any task with > 5 independent sub-steps
Decomposition Template
## Task: [name]
### Sub-task 1: [specific deliverable]
### Sub-task 2: [specific deliverable]
### Sub-task 3: [specific deliverable]
### Merge & Verify
Each sub-task becomes a delegate_task() call with its own context.
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.
- 5d ago First seen · 57 lines · 20 tokens per session scan A f74d76a70512
context-aware-task-decomposition is a skill published in the GitHub repository ChrisLamDev/hermes-core-skills (9 stars, last pushed 2mo ago), licensed MIT. It adds 20 tokens to every session and 345 once invoked, about $0.0001 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 skills, from other repositories
author
Scaffold a conforming AI-layer artifact — a rule, skill, hook, agent, or output style — that already passes the format checks, then verify it. Use when adding any new AI-layer artifact — a shared one under .agents/ or a Claude-only one under .claude/ — so it matches the existing shape instead of drifting.
close
The "wrap" verb of the loop — end the session by writing the journal entry, updating touched project READMEs, committing, and pushing. Auto-invoked as the tail of /spec and /implement (the producer verbs) on clean completion, so "wrap" is not a verb the user types; also runnable directly. Dispatches /implement-audit…
explain
Deep research into a topic or issue — investigate until confident, then present executive summary and root cause analysis. Use when you need to understand WHY.
spec-audit
Adversarial review of a freshly-written or materially-rewritten SPEC, before any code is written. Runs an independent reviewer against the SPEC's design and the spirit-check agent against the user's verbatim ask, in one pass. Auto-fired by /spec; also callable standalone after any SPEC edit. Emits the spec-audit…
debate
Adversarial debate — spawn competing AI agents to argue a question, then synthesize the strongest conclusion. Use when the user says "debate this", "I'm torn between X and Y", "red-team this plan", or "council on this".
implement-audit
Adversarial review of freshly-built code — find what was missed, what's inconsistent, what breaks. The single code reviewer for the loop, and it runs exactly once per session. Auto-fired by /implement; standalone-callable after any substantial ad-hoc change. Emits the implement-audit: commit trailer the git hook…