Installs into .claude/skills of the current project.
Sponsors
SSkills DirectorySkills Directory
Know which skills are safe — weekly.
SSkills DirectorySkills Directory
Is that skill safe to install?
Are you the author of Context?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/buchbend-context)
---
name: lore:context
description: Show the full timeline of context Lore injected this session.
Pure cache read — no live I/O, no re-gather. Run with "/lore:context".
user_invocable: true
---
# Context — show what Lore injected this session
One Bash call, then one text message:
1. Run `lore hook context-log` with `dangerouslyDisableSandbox: true`
(it walks `/proc` for the Claude Code ancestor PID; sandboxed
seccomp blocks that on the first try).
2. Output the captured stdout **verbatim** inside a fenced code block.
No preamble, no commentary — Bash output is collapsed in the UI by
default, so echoing it as text is what makes the timeline visible.
If the log is empty, the command prints a message; relay it as-is.
Mention `lore setup` only if the user asks why.
## Do not
- Re-narrate the output.
- Invoke `lore hook session-start` to re-generate — use `lore_search` or
`lore_context_pack` for a fresh gather.
- Read wiki notes, grep the catalog, or call MCP tools.
## Related
- `lore_context_pack` — fresh, scoped gather (different from showing the cache)
- `lore off` — mute hooks for the session
