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 commands/vladm3105/aidoc-flow-framework/feedbackgit clone --depth 1 https://github.com/vladm3105/aidoc-flow-frameworkWrote 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/commands/vladm3105/aidoc-flow-framework/feedback)<a href="https://agentmods.dev/commands/vladm3105/aidoc-flow-framework/feedback"><img src="https://agentmods.dev/badge/commands/vladm3105/aidoc-flow-framework/feedback.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 | $0.00046 | $0.01468 |
| Opus 5 | $0.00023 | $0.00734 |
| Sonnet 5 | $0.00009 | $0.00294 |
| Haiku 4.5 | $0.00005 | $0.00147 |
Grade A, and why
feedback 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 2d 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 — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Feedback
Separate channel from /aidoc-flow:bug-report. Bugs are "something is broken";
feedback is "what would you change" or "what worked." Same machinery — the LLM
drafts a structured issue from the user's one-line prompt + conversation
context, the URL is prefilled, the user clicks Submit on github.com.
Invocation
/aidoc-flow:feedback <one-sentence summary of what worked / didn't / what to change>
Examples:
/aidoc-flow:feedback the help output is too long on small terminals
/aidoc-flow:feedback love that /status shows last edit date, big help
/aidoc-flow:feedback could /budget min also skip the audit step
Running /aidoc-flow:feedback with no text still works — the LLM uses only
the conversation context.
Instructions
-
Capture the user's argument — every word after
/aidoc-flow:feedbackis the user's report. Store it asuser_remark. If empty, setuser_remark = "(none provided — using conversation context only)". -
Gather the version stamp — read:
${CLAUDE_PLUGIN_ROOT}/VERSION→plugin_version${CLAUDE_PLUGIN_ROOT}/FRAMEWORK_SPEC_VERSION→framework_spec
-
Read the conversation context — review the recent messages. Look for:
- Which command, skill, or layer the user was working with when the
remark applies (e.g.
/aidoc-flow:help,doc-brd-autopilot, layerBRD). - Whether the remark is a feature idea, praise, friction, or a question. Pick exactly one for the issue's lead-in.
- Any concrete suggestion the user implied (e.g. "could /budget min
also skip the audit step" → suggestion: "extend
budget: minprofile to skip optional audit passes").
- Which command, skill, or layer the user was working with when the
remark applies (e.g.
-
Draft the issue. Produce
titleandbody. The body matches the structure in.github/ISSUE_TEMPLATE/feedback.md.Title rules:
- One line, ≤ 80 characters.
- Lead with the category in brackets when known:
[idea],[praise],[friction],[question]. If you cannot classify confidently, omit the bracket. - Concrete and specific. Reuse the user's wording when it is specific; rephrase when vague.
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.
- 2d ago First seen · 169 lines · 46 tokens per session scan A 780bdb144702
feedback is a command published in the GitHub repository vladm3105/aidoc-flow-framework (17 stars, last pushed 2d ago), licensed MIT. It adds 46 tokens to every session and 1,468 once invoked, about $0.0002 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 commands, from other repositories
create-command
Create a new obsidian-second-brain command via interview - zero markdown editing required.
obsidian-health
Run a vault health check - grouped by severity, detects contradictions, concept gaps, stale claims, and structural issues.
obsidian-visualize
Generate a visual canvas map of your vault - see the shape of your second brain and how knowledge connects.
obsidian-architect
Scan a codebase and write a maintained set of architecture notes into the vault - overview, per-module notes, key decisions. Re-run to refresh without clobbering your edits.
obsidian-export
Export a clean structured snapshot of the vault that any agent or tool can consume - flat JSON, markdown index, or an OKF (Open Knowledge Format) bundle.
obsidian-retrieval-eval
Measure how well vault search finds the right note for a natural-language question - recall@k and MRR, with the concrete failures.