/context why <id> now shows full memory object details: parent,
children count, hash prefix, title from attributes.
/context dropped replaced literal stub with computed estimate of
pruned messages based on token budget (msg_count * 60 vs 8192).
- channel-tui-main: enhanced both debug handlers
- TUI Main: 102/102
audit-verify-hash counts total objects and those with missing/empty
hashes. /audit verify uses it to report VERIFY PASS or MISSING
HASHES count. fboundp-guarded.
- core-memory: audit-verify-hash fn, 1 test
- channel-tui-main: updated /audit verify handler
- Core: 90/90 TUI Main: 102/102
*tag-trigger-count* hash table tracks per-session tag triggers.
tag-trigger-record increments count, called from
dispatcher-privacy-severity on each matched tag. /tags shows
trigger count per tag.
tag-categories-load now falls back to PRIVACY_FILTER_TAGS env var
when TAG_CATEGORIES is not set (backward compat). All entries
default to :block severity.
- security-dispatcher: *tag-trigger-count*, tag-trigger-record,
updated tag-categories-load, wired dispatcher-privacy-severity
+2 tests (trigger record, privacy fallback)
- channel-tui-main: /tags shows trigger counts
- Core: 88/88 TUI Main: 102/102
/context now shows full budget breakdown: IDENTITY, TOOLS,
TIME+CONFIG, LOGS with per-section token estimates, visual bar
chart, and percentage used. Over-80% warning.
Estimates computed from live state: identity/config lengths, tool
registry count, message count. Budget cap at 8192 tokens.
- channel-tui-main: rewritten /context handler, 1 new test
- TUI Main: 101/101
test-on-key-help checked for /eval in help output. The rewritten
help list dropped /eval which is still a working command. Fixed by
adding it as the first entry.
TUI Main: 98/98
Search mode activated by /search <query>. State fields: :search-mode,
:search-query, :search-matches, :search-match-idx. Up/Down arrows
navigate between matches, Enter jumps to current match, Escape exits.
search-highlight wraps matching substrings in **bold** for markdown
rendering. View-chat shows search header bar with match count and
current position.
- channel-tui-state: 4 search state fields in init-state
- channel-tui-main: modified /search handler, search-mode key handlers
(Up/Down/Enter/Escape), 3 new tests (activate, escape, nav)
- channel-tui-view: search-highlight fn, search header bar,
highlighted content in count+render loops
- TUI Main: 97/98 (1 pre-existing flake) View: 29/29
resolve-hitl-panel marks the most recent panel message with
:panel-resolved (:approved or :denied) and writes back to
the message vector. View-chat renders resolved panels with
dimmed color instead of :hitl theme color.
/approve and /deny handlers call resolve-hitl-panel after
sending structured events to the daemon. Confirmation messages
now use checkmark/crossmark prefixes.
- channel-tui-main: resolve-hitl-panel fn, wired into handlers
- channel-tui-view: is-resolved check for panel dimming
- +2 tests: panel-after-approve, panel-after-deny
- TUI Main: 88/89 (1 pre-existing flake)
View-chat renders gate-trace-lines as colored dim lines below agent
messages. Ctrl+G toggles collapse per message (stored in
:collapsed-gates state field). Default: visible. /why shows last
gate trace as system messages.
Tab integration deferred (paren fragility in on-key cond default
case). Ctrl+G is functionally equivalent for toggle UX.
View: 29/29 TUI Main: 85/86
call-with-tool-timeout wraps tool execution with sb-ext:with-timeout
using per-tool timeout from *tool-timeouts*. On timeout returns
(:status :error :message "Timed out after Ns"). Wired into
action-tool-execute before the funcall. Timeout result detected and
propagated as :tool-error.
- core-act: call-with-tool-timeout fn, wired into action-tool-execute
- Act tests: +3 (timeout enforcement test)
- Core: 88/88
/search <query>: case-insensitive substring search across message
history. Reports match count, previews with context around matches.
/context: shows message count, focus, token estimate, last 5 messages.
- channel-tui-main: /search and /context handlers, 1 test each
- TUI Main: 85/86 (1 pre-existing core flake)
Gate trace lines rendered below each agent message in dim color.
Collapsed-gates state field for Tab toggle (default: visible).
Uses passepartout::gate-trace-lines for colored entries.
- channel-tui-view: view-chat renders gate-trace after message content
- channel-tui-state: :collapsed-gates field in init-state
- View tests: 29/29 (1 new state-field test)
- CONFIG section in system prompt: providers, context window, gate count,
rules learned, docs path
- /why TUI command: shows most recent gate trace from message history
- assemble-config-section reads live state at each think() call
- Core: 75/76 TUI Main: 77/78 (1 pre-existing RCE test flake)
Live config section injected into system prompt between time and
IDENTITY. assemble-config-section reads *provider-cascade*,
tokenizer-context-limit, gate count, and *hitl-pending* at each
think() call. fboundp-guarded. Tested.
- core-reason: assemble-config-section fn, config-section binding,
injected into all 3 prompt assembly paths
- Reason tests: +4 checks (Passepartout, version, gates)
char-width: contract 5, 4 tests (6 assertions), 100% pass
ASCII=1, CJK/Hangul/Kana/halfwidth=2, combining marks=0, tab=8
Pure Lisp, ~25 lines, no deps. Used by word-wrap for unicode.
status bar: contract 6, timestamp right-aligned at (- w 12)
Fixes overlap where focus map and timestamp both drew at :y 2 :x 1
RED: embedding-backend-native does not exist. No CFFI llama binding.
GREEN (REPL progress):
- cffi:define-foreign-library libllama → loaded
- defcstruct with correct sizes (verified via C sizeof program):
llama-mparams (72 bytes), llama-cparams (136 bytes), llama-batch (56)
- Field offsets verified via C offsetof program
- llama_backend_init discovered as required prerequisite
- llama-model-default-params correctly fills 72-byte struct (verified)
- llama-embedding CLI verified: 768-dim vectors, 22ms/4tokens
BLOCKED: llama_model_load_from_file segfaults via CFFI. Suspect struct-by-value
vs pointer ABI mismatch on x86-64. Needs interactive SBCL REPL to debug the
calling convention (structs >16 bytes passed by hidden reference on SysV).
CFFI bindings preserved in org/system-model-embedding-native.org for
continued REPL work. Includes: model load, context create, tokenize,
encode, embeddings-ith, batch init/free.
Model: nomic-embed-text-v1.5.Q4_K_M.gguf (80MB, 768-dim, nomic-bert)
at ~/.local/share/passepartout/models/
Adds Discord gateway: REST API POST /channels/{id}/messages for
sending, HTTP GET for polling messages. Maps Discord mentions to
:user-input signals. HITL commands intercepted before injection.
Adds Slack gateway: Web API chat.postMessage for sending,
conversations.history for polling. Uses SLACK_TOKEN from vault.
Each gateway registered in *gateway-registry* following the same
jail-loaded skill pattern as Telegram and Signal.
Registry now has 4 platforms: telegram, signal, discord, slack.
RED: Messaging suite had only 1 test (5 checks). No Telegram or Signal
integration tests existed.
GREEN: 4 new tests, 12 new checks (5 → 17):
test-telegram-send-format: verifies URL/body construction for
telegram-send — URL contains sendMessage + token, body encodes
chat_id + text as JSON.
test-telegram-poll-hits-interception: verifies HITL commands
(/approve, /deny, /approve <token>) are intercepted before
signal injection. Non-HITL messages pass through.
test-signal-send-format: verifies signal-send constructs correct
CLI args for signal-cli (account, send, -m, text, chat-id).
test-signal-poll-json-parse: verifies signal-cli JSON output is
parsed correctly — extracts envelope source and dataMessage text.
Test: 123/0 across 13 suites (messaging 17/0).
