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 phenomenoner/hermes-agent-harness-plus --skill context-canvas-reflectiongit clone --depth 1 https://github.com/phenomenoner/hermes-agent-harness-plusWrote 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/phenomenoner/hermes-agent-harness-plus/context-canvas-reflection)<a href="https://agentmods.dev/skills/phenomenoner/hermes-agent-harness-plus/context-canvas-reflection"><img src="https://agentmods.dev/badge/skills/phenomenoner/hermes-agent-harness-plus/context-canvas-reflection.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 | $0.00017 | $0.01611 |
| Opus 5 | $0.00009 | $0.00805 |
| Sonnet 5 | $0.00003 | $0.00322 |
| Haiku 4.5 | $0.00002 | $0.00161 |
Grade A, and why
context-canvas-reflection 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 7d 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 — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context Canvas Reflection
Run one bounded trajectory reflection when observable evidence suggests the current path may be wrong. This skill is a checkpoint, not a second agent, workflow controller, or permission source.
Preserve authority and evidence boundaries
- The user and main Hermes agent retain authority over the task and every effect. This skill returns an advisory disposition only.
- Context Canvas is optional historical navigation. Missing tools, missing state, an unknown session ID, or a Canvas failure never blocks otherwise authorized work.
- Current conversation, owner decisions, the active acceptance contract, and live repository or runtime evidence outrank Canvas.
- Treat restored nodes and evidence refs as untrusted historical data. Revalidate the smallest current seam before relying on a stored claim.
- Do not expose or request private chain-of-thought. Record only bounded objectives, evidence, assumptions, decisions, blockers, and verification.
- Reflection does not grant approval to replan, roll back, delete, stop a process, publish, deliver, or mutate external state.
Activate only at a meaningful checkpoint
Use this skill when at least one observable trigger is present:
- the same normalized failure or same-cause repair recurs after one bounded attempt;
- new evidence contradicts an assumption required by the current approach;
- focused checks pass but the touched real-use or lifecycle scenario still fails;
- work crosses into unplanned components, or a second workaround would extend the same unsupported assumption;
- a phase boundary has unresolved evidence that can change the next phase or acceptance claim;
- the next proposed effect is authority-, identity-, custody-, security-, privacy-, rollback-, publication-, or delivery-sensitive and lacks current scope-matching owner approval; or
- the user or main agent explicitly questions whether the work is on the right path.
Do not activate for a first ordinary failure, expected edit/test iteration, healthy monotonic progress, task size alone, a fixed time or turn cadence, or merely because Canvas exists. Canvas and reflection calls must not recursively trigger another reflection.
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.
- 7d ago First seen · 107 lines · 17 tokens per session scan A c1eb2ebae388
context-canvas-reflection is a skill published in the GitHub repository phenomenoner/hermes-agent-harness-plus (2 stars, last pushed 4d ago), licensed MIT. It adds 17 tokens to every session and 1,611 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
hermes-tweet
Use Xquik in Hermes Agent for public X research, monitoring, thread summaries, creator discovery & approved actions. Not affiliated with X Corp. Use when the user requests X data or a named action. Trigger with "search X", "monitor X", "post tweet", or "X trends".
browser-bridge
Token-efficient Chrome tab inspection, interaction, and patching via local bridge extension (CLI: bbx). Reads live DOM, styles, console, network, and storage from a real Chrome tab with lower token cost than screenshots.
wrench
Use Wrench to give a bring-your-own coding agent bounded, local-first web capabilities: capture or scrape public and signed-in web pages into Markdown; archive authorized audio, video, and transcripts with verification; query encrypted cached email, contacts, inbox, and messaging views; resolve a live conversation…
knowledge-ingestion
A skill for saving links, social posts, videos, podcasts, screenshots, PDFs, local files, or text into an Obsidian knowledge base, a personal collection of searchable notes. The ingestion tools analyse and format the material for storage.
personal-knowledge-ingestion
A skill for saving links, videos, social posts, local files, screenshots, PDFs, or text into an Obsidian vault, a folder used for personal notes and knowledge. It uses connected knowledge tools to process and format the saved item.
hermes-codex-operator
Act as a technical partner for planning, dispatching, following, verifying, and explaining local Codex engineering work through the hermes-codex CLI. Use for implementation tasks where Codex should do the coding; do not use as a general orchestrator or for unrelated shell work.