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.
git clone --depth 1 https://github.com/Cfvillarroel/carryoverWrote 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/cfvillarroel/carryover/headroom)<a href="https://agentmods.dev/commands/cfvillarroel/carryover/headroom"><img src="https://agentmods.dev/badge/commands/cfvillarroel/carryover/headroom/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/commands/cfvillarroel/carryover/headroom"><img src="https://agentmods.dev/badge/commands/cfvillarroel/carryover/headroom.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 | $0.00011 | $0.00276 |
| Opus 5 | $0.00005 | $0.00138 |
| Sonnet 5 | $0.00002 | $0.00055 |
| Haiku 4.5 | $0.00001 | $0.00028 |
Grade A, and why
headroom scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
You are the headroom operator. Run these commands and summarize the result clearly and briefly:
HR=~/.headroom/venv/bin/headroom
DB="$HOME/.headroom/memory.db" # the proxy's real store (NOT the ./headroom_memory.db in cwd)
$HR install status # proxy: running/healthy, port
$HR memory stats --db-path "$DB" # total memories and by scope
$HR memory list --db-path "$DB" --limit 10 # latest memories
$HR output-savings 2>/dev/null || true # tokens saved (if data is available)
Then report:
- Whether the proxy is up and healthy (or down → suggest
hr-on). - How many memories there are (USER scope = shared across repos).
- The token savings if available.
If the user passed arguments in $ARGUMENTS, interpret them as a subcommand of
headroom memory (e.g. show <id>, list --scope USER, --since 7d) and run it
always adding --db-path "$HOME/.headroom/memory.db" instead of the default summary.
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.
- 9d ago First seen · 24 lines · 11 tokens per session scan A f43c6c1d2c2a
headroom is a command published in the GitHub repository Cfvillarroel/carryover (5 stars, last pushed 24d ago), licensed MIT. It adds 11 tokens to every session and 276 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 commands, from other repositories
context-restore
Expert Context Restoration Specialist focused on intelligent, semantic-aware context retrieval and reconstruction across complex multi-agent AI workflows. Specializes in preserving and reconstructing project knowledge with high fidelity and minimal information loss.
context-save
An elite context engineering specialist focused on comprehensive, semantic, and dynamically adaptable context preservation across AI workflows. This tool orchestrates advanced context capture, serialization, and retrieval strategies to maintain institutional knowledge and enable seamless multi-session collaboration.
obsidian-catchup
Review and process everything captured on the go from the Telegram journal bot - voice, text, images, PDFs, links - waiting in the catchup queue. You pull it when you are back at the laptop; nothing is processed autonomously.
recall-loop
Periodic memory digest — designed for use with Claude Code's /loop scheduler. Shows recent memories and pending to-dos in a compact, scannable format.
init-workspace-flow-shells
Phase 2 Shells of init-workspace-flow.
context-stats
Display context window usage and token statistics.