Skip to main content

Class: RetrievalFeedbackSignal

Defined in: packages/agentos/src/memory/feedback/RetrievalFeedbackSignal.ts:112

Detects which injected memory traces were used vs ignored by the LLM, persists those signals to the retrieval_feedback table, and applies a best-effort trace-strength update in memory_traces.

Lifecycle:

  1. Before generation: retrieve relevant traces and inject them into the prompt.
  2. After response delivery (non-blocking): call detect(injectedTraces, response).
  3. The signal is recorded immediately and the underlying trace is nudged toward reinforcement or decay.
  4. The consolidation pipeline can still read getStats(traceId) later for broader aggregate decisions.

Constructors​

Constructor​

new RetrievalFeedbackSignal(brain, similarityFn?): RetrievalFeedbackSignal

Defined in: packages/agentos/src/memory/feedback/RetrievalFeedbackSignal.ts:120

Parameters​

brain​

SqliteBrain

The agent's SQLite brain; used to persist and query feedback rows.

similarityFn?​

(a, b) => Promise<number>

Optional semantic similarity function for higher-fidelity detection. Receives two strings and returns a promise of a similarity score in [0, 1]. When provided, the score supplements the keyword heuristic, but the current implementation uses the keyword path only (reserved for future use).

Returns​

RetrievalFeedbackSignal

Methods​

detect()​

detect(injectedTraces, response): Promise<RetrievalFeedback[]>

Defined in: packages/agentos/src/memory/feedback/RetrievalFeedbackSignal.ts:150

Detect which of the injected traces were referenced in response, persist the signals to retrieval_feedback, update the corresponding memory_traces rows, and return the full feedback array.

Keyword heuristic:

  • Extract all words > 4 characters from each trace's content field, lowercased and stripped of non-alphanumeric characters.
  • Compute matchRatio = (words found in response) / (unique keywords).
  • Signal = 'used' if matchRatio > 0.30, else 'ignored'.

When a trace has no qualifying keywords (all words ≤ 4 characters), it is treated as 'ignored' — there is nothing to match against.

Note: detect() is synchronous under the hood (better-sqlite3) but declared async to allow future replacement with an LLM-backed detector.

Parameters​

injectedTraces​

MemoryTrace[]

Memory traces that were injected into the prompt.

response​

string

The LLM's generated response text.

Returns​

Promise<RetrievalFeedback[]>

Array of RetrievalFeedback events, one per injected trace.

getHistory()​

getHistory(traceId, limit?): RetrievalFeedback[]

Defined in: packages/agentos/src/memory/feedback/RetrievalFeedbackSignal.ts:259

Retrieve the feedback history for a single trace, ordered by most-recent first.

Parameters​

traceId​

string

The memory trace ID to look up.

limit?​

number = 100

Maximum number of rows to return. Defaults to 100.

Returns​

RetrievalFeedback[]

Array of RetrievalFeedback events, most-recent first.

getStats()​

getStats(traceId): object

Defined in: packages/agentos/src/memory/feedback/RetrievalFeedbackSignal.ts:287

Return aggregate counts of 'used' and 'ignored' signals for a trace.

Useful for the consolidation pipeline to decide whether to apply penalizeUnused (many ignores) or updateOnRetrieval (many used).

Parameters​

traceId​

string

The memory trace ID to aggregate.

Returns​

object

{ used, ignored } counts.

ignored​

ignored: number

used​

used: number