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/projectbluefin/testsuite/opsnpx skills add projectbluefin/testsuite --skill opsgit clone --depth 1 https://github.com/projectbluefin/testsuiteWhat 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 | $0.00038 | $0.02318 |
| Opus 5 | $0.00019 | $0.01159 |
| Sonnet 5 | $0.00008 | $0.00464 |
| Haiku 4.5 | $0.00004 | $0.00232 |
Grade B, and why
ops scanned grade B with 1 finding 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- Calling `sudo podman load` instead of rootless load (image goes to root storage) The source is not reproduced here
No licence file
A repository with no LICENSE is all rights reserved by default, so the body is not copied here. The metadata, the measurements and the link are.
What ships with it
43 files 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.
- references/arc-ghost-runners-local-dev-routing.md 1.9 KB
- references/argo-mutex.md 1018 B
- references/bazzite-extension-state-use-getextensioninfo-not-shell-eval.md 1.5 KB
- references/before-scenario-and-after-scenario-setup-guard-pattern.md 1.3 KB
- references/bootc-install-creates-0-origin-alongside-0-deploy-find-must.md 1.5 KB
- references/bootloader-flag-requires-bootc-0-1-13.md 583 B
- references/common-shell-tools-missing-on-bluefin-lts-bluefin-gdx.md 772 B
- references/common-suite-execution-model-runner-container-not-inside-vm.md 1.5 KB
- references/composefs-file-capability-regression.md 1.3 KB
- references/composite-actions-vs-checkout-for-cross-repo-scripts.md 1.4 KB
- references/containerfile-runner-requirements.md 2.2 KB
- references/coverage-snapshot-generated.md 2.2 KB
- references/cross-image-tag-skipping-bluefin-on-non-bluefin-images.md 1.2 KB
- references/do-not-add-a-second-monitor-flag-to-qemu.md 650 B
- references/fedora-version-targets.md 964 B
- references/gdm-autologin.md 1.1 KB
- references/ghost-lab-poller-failures.md 2.7 KB
- references/git-restore-vs-git-checkout-for-full-directory-reset.md 1.2 KB
- references/github-api-rulesets-require-put-not-patch.md 912 B
- references/gnome-50-requires-qecore-4-12.md 504 B
- references/gnome-shell-readiness-failures.md 2.9 KB
- references/nvidia-services-always-fail-in-qemu.md 436 B
- references/oci-layer-caching.md 672 B
- references/oneshot-systemd-service-state-use-result-not-activestate.md 1.4 KB
- references/oras-pull-always-use-a-directory-never-a-file-path.md 881 B
- references/polkit-rules-path-check-both-directories.md 843 B
- references/python-3-14-sys-executable-is-empty-in-pid-host-containers.md 756 B
- references/qecore-headless-restarts-gdm-bus-socket-churn.md 3.9 KB
- references/quarantine-tag-enforcement-two-layers-required.md 1.0 KB
- references/results-json-captures-first-pass-only.md 549 B
- references/rootless-podman-load-fails-in-vm-exit-125.md 1.1 KB
- references/smoke-suite-pre-existing-lab-failures-gnome-50-at-spi.md 2.3 KB
- references/smoke-suite-run-vs-run-host-for-network-checks.md 1015 B
- references/ssh-step-timeout.md 414 B
- references/stdin-heredoc-consumption-bug-in-run-gnome-tests-yaml.md 1.1 KB
- references/sys-exit-1-in-before-scenario-kills-behave.md 658 B
- references/systemd-oomd.md 536 B
- references/ublue-motd-prepended-to-ssh-output.md 755 B
- references/user-bootstrap-primary-group-bug-on-fresh-boots.md 762 B
- references/workflows-without-a-pull-request-trigger-break-main-silently.md 2.3 KB
- references/xdg-desktop-portal-and-graphical-session-target.md 1.4 KB
- references/xdg-session-type-and-xdg-session-desktop-must-be-forwarded.md 893 B
- references/yaml-orphan-keys-in-e2e-yml.md 762 B
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 · 134 lines · 38 tokens per session scan B b765dac7c168
ops is a skill published in the GitHub repository projectbluefin/testsuite (2 stars, last pushed 2d ago), with no licence file. It adds 38 tokens to every session and 2,318 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
eng-runbook
服务概述 + alerts 表 + dashboards + 操作命令 + on-call + 事故清单.
lastlight-debug
Debug a running Last Light instance over its admin API with the lastlight CLI — no SSH. Use when the user wants to "look at / debug a failed or stuck workflow", "why didn't the bot do X", "check a bad agent run / phase error on ", "tail a session", "search the logs", "read the server/docker logs", "list/trigger a…
dpf-worktree-per-session
Use when starting, entering, auditing, or managing a concurrent DPF coding session that touches the working tree.
debug-production
Debug a running Last Light instance (usually production) via the lastlight CLI instead of SSH. Use when investigating a failed/stuck workflow, a bad agent run, a phase error, or "why didn't the bot do X" — anything that previously meant SSHing in to read SQLite or session logs.
dpf-file-backlog-item
Use when new DPF work needs to enter the backlog — a feature gap, bug, tool, skill or doc gap, automated detection, or user request.
incident-postmortem
Use after an incident is resolved — drafts a blameless postmortem from timeline notes, alerts, and chat threads.